Saleem Bahaj, Angus Foulis, Gabor Pinter and Paolo Surico
Changes in interest rates affect different parts of the economy differently. In this post, building on a recent working paper, we consider how different types of firms respond to interest rate changes. We focus on firm level employment and ask which firms do the most hiring and firing when monetary policy adjusts. For instance, how important is the age of the firm, its balance sheet position or its size in determining the firm level response to interest rates? Furthermore, do these patterns of responses tell us something about how monetary policy affects the economy?
We combine data from Companies House on the employment of individual firms in the UK with a measure of surprise changes in interest rates. The latter measure is important. Interest rates rise and fall with the business cycle. Our focus is not on which types of firms are more cyclical but rather which are more sensitive to interest rates. We need to isolate interest rate movements unrelated to economic developments. Luckily, existing work can help: we use changes in the price of interest rate futures traded in financial markets in extremely tight windows around monetary policy announcements to get at the surprise, random component of an interest rate decision.
Our firm-level sample contains information on large, listed firms and small private companies. It covers firms that have been active for over a century and those who have just been born. The data also contains information on the balance sheet position of individual firms, so we can observe how leveraged a firm is; that is how much of the firm’s assets are funded through borrowing from creditors rather than money invested by the owners. In short, it has good coverage of the different parts of UK PLC.
Younger, more indebted firms cut employment the most when interest rates rise
Figure 1 shows the relationship between employment and interest changes for different firms along the dimensions of size, age and firm leverage. Imagine a surprise announcement by the Bank of England that it would raise interest rates such that for the next year they would be 0.25% higher on average. The figure shows how much, in total, these different types of firms would have cut their employment by 2 years later.
Figure 1: Employment response, after 2 years, to a 25bp interest rate increase
a) Age
b) Leverage
c) Size
This sensitivity of firm employment to a change in interest rates: (i) is decreasing in the age of the firm – a very young firm will cut its employment by 2% (assuming it survives), whereas firms older than 30 barely react; (ii) depends on the firm’s leverage – a firm in the upper three quintiles of the leverage distribution (firms who have debt worth more than half their assets) sheds around 1.5% of its workers; there is little reaction from firms who are less debt dependent; (iii) has a less than clear cut relationship with firm size – smaller and larger firms appear less sensitive than those in the middle of the size distribution; moreover among firms with less than 500 employees, which make up the majority of aggregate employment, the responses are similar.
Digging deeper, Figure 2 shows what happens when we sort by age and leverage at the same time. It presents the dynamic response (after 1, 2, 3 years etc.) to the interest rate change rather than just at the 2-year horizon. This figure confirms that both age and leverage matter. Among old firms (aged more than 15 years) highly levered firms (above the median) are more sensitive than firms with less debt. Among levered firms the young are more sensitive than the old. And younger-more levered firms (the top left quadrant in Figure 2) are the most sensitive of them all. This is interesting from an aggregate perspective. Only a quarter of employees work at firms that are both young and highly levered. However, because the sensitivity of those firms to interest rate changes is so large, our findings suggest the response of this group explains more than half the total fall in employment that follows an increase in interest rates.
Figure 2: Employment responses by age and leverage
Difficulties in accessing finance explain why these firms behave differently.
This leads us to the second part of our question: why are these types of firms so sensitive to interest rates? Highly levered firms and young firms have something in common: they both tend to be reliant on borrowing and, potentially, face difficulties getting more finance. Highly levered firms have already used up their debt capacity. Whereas young firms tend to need finance to grow, have less history for banks to lend against, and tend to be more risky prospects reliant on a few key personnel. As a result, younger, more levered firms may need to pledge other assets as collateral to obtain the funding they need.
This aligns with an influential theory in macroeconomics: “the financial accelerator”. Higher interest rates lower asset prices, this reduces the value of the assets that firms that are constrained financially borrow against which means these firms are even less able to access credit. The firms then cut back on workers and investment more than they otherwise would have, which brings us back to the start of this post.
Isolating this effect is challenging. The most common form of collateral supporting corporate credit in the UK is real estate: 75% of loans are secured against it. Real estate prices are very sensitive to interest rates. Moreover, real estate prices respond differently to interest rates in different parts of the UK. However, real estate prices in the firm’s region may also affect demand for the firm’s goods: higher house prices may lead to equity extraction or less saving, boosting consumer spending. This blurs differences between firms’ access to finance and the market conditions they face.
To overcome this issue, we make use of some prior work. While real estate is a common source of collateral, it is not just the firm’s buildings that are used but also the housing of the firm’s directors. Around half of loans to firms have personal assets listed as security, rising to around two thirds for loans to younger and more levered firms. Crucially, directors must declare their usual residential address to Companies House so we can observe where they live. For two thirds of directors, this is a different region to their firm.
This means that we can compare two young, highly levered firms operating in the same region (so facing the same local demand conditions). However, one firm has a director who lives in a region where house prices are very sensitive to interest rates (i.e. the price of a potential source of collateral will move a lot) and the other firm has a director whose house value will not change much. We can repeat this for older, less levered firms. We refer to the sensitivity of director house value to interest rates as the “director beta”. This can be measured on a region by region basis using regional house price data and our interest rate surprise measure.
Bringing all these pieces together, Figure 3 summarises our results. It shows the response of employment for different groups of firms relative to older, less levered, low director beta firms (relative results are required for technical reasons). The results are consistent with Figure 2, but also show that among the younger, more levered group, firms with a high beta (i.e. more reactive collateral values) adjust employment much more than the low beta firms. This is not the case for older, less levered firms. Director beta makes no difference. This makes sense. Older, less levered firms are less likely to have difficulties accessing finance so fluctuations in collateral values are less important.
Figure 3: Employment responses by director beta (relative to old, low leverage, low beta firms)
What then is the upshot of this analysis? There are three things we can conclude. First, interest rate changes have very different effects on different types of firms in the UK. And at least part of the reason for these differences is that firms are heterogeneous in the ease with which they can access finance. Second, interest rates affect the economy through changing asset prices and collateral values, which alters firms’ access to credit. Third, a key collateralisable asset for firms is the housing of directors and monetary policy works in part by shifting its value.
Saleem Bahaj works in the Bank’s Research Hub, Angus Foulis and Gabor Pinter work in the Bank’s Centre for Central Banking Studies and Paolo Surico works at the London Business School.
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.
