Miguel García-Posada and Sergio Mayordomo
In February, the Financial institution hosted its inaugural Financial institution of England Agenda for Analysis (BEAR) convention, with the theme of ‘The Financial Toolkit’. As a part of our occasional sequence of Visitor Posts by exterior presenters at Financial institution analysis occasions, the authors of 1 paper from the BEAR convention define their findings on the impact of damaging charges on Spanish banks…
Over the previous decade, a number of central banks have lower coverage charges under zero. These are unlikely to work in the identical vogue as price cuts in optimistic territory, due to difficulties with charging damaging rates of interest on retail deposits, and due to banks’ damaging return on extra liquidity. In a latest paper we discover the impact on financial institution lending by combining information on publicity to damaging charges with banks’ steadiness sheets, the Spanish credit score register and companies’ steadiness sheets. We discover it was solely after a number of years of damaging charges, when deposit charges reached zero, that affected banks started to lower credit score provide and elevated mortgage charges. This impact was particularly sturdy for undercapitalised banks and lending to dangerous companies.
Why would possibly damaging charges work otherwise?
Financial principle suggests the impact on financial institution lending of price cuts in damaging territory may go both approach. One the one hand, it’s nicely documented that the existence of money instead retailer of worth means it could be tough for banks to set damaging charges on deposits (eg Heider et al (2019); Eggertsson et al (2020)). This squeezes the web curiosity earnings of banks with a excessive deposit share: retail deposit charges are caught at zero, whereas the remainder of liabilities (eg wholesale funding) and belongings (eg floating-rate loans) are repriced at a decrease degree.
Alternatively, as a result of banks with extra liquidity earn a damaging return, they’ve incentives to extend their lending to the non-public non-financial sector in a bid to cut back their extra liquidity holdings (eg Basten and Mariathasan (2018); Demiralp et al (2019)). This portfolio-rebalancing channel might suggest greater threat taking, as risk-free extra liquidity is transformed into financial institution lending.
The evaluation of the impression of damaging rates of interest on banks’ credit score provide and threat taking is prone to rely on the time over which coverage charges are damaging. A ‘negative-for-long’ situation would possibly look fairly totally different to a shorter spell of charges under zero. As documented by Eggertsson et al (2020), damaging charges might have contractionary results solely when retail deposit charges attain the zero decrease certain (ZLB).
Estimating the impact of damaging charges on Spanish banks
We examine the impact of the ECB’s damaging deposit facility price (DFR) on the provision of credit score by Spanish banks to non-financial companies throughout a protracted time interval, 2014–19. Spain gives a great case examine right here due to publicity to damaging coverage charges, mixed with a pretty dataset which is right for figuring out the results of damaging charges on financial institution lending. As well as, deposit charges in Spain have been nicely above the ZLB when damaging rates of interest have been launched within the euro space and it took them a number of years to succeed in the ZLB, which gives a really perfect alternative to check the results of a ‘negative-for-long’ situation on credit score provide. To do that, we construct a singular dataset that contains the universe of loans granted to Spanish companies from the Credit score Register of the Financial institution of Spain, banks’ and companies’ steadiness sheets and confidential survey information from the ECB’s Financial institution Lending Survey (BLS).
The BLS is a quarterly survey via which euro space banks are requested about developments of their respective credit score markets since 2003 (for extra detailed info see Köhler-Ulbrich, Hempell and Scopel (2016)). Our identification technique exploits heterogeneous publicity to damaging charges throughout banks. Particularly, we estimate an ‘affected dummy’, which is predicated on the likelihood {that a} financial institution is adversely affected by the damaging rates of interest (affected for brief), primarily based on the confidential solutions to the BLS. We assume {that a} financial institution is affected, and our ‘affected dummy’ equals 1, if the likelihood that it studies that the ECB’s damaging DFR contributed to a decline in its internet curiosity earnings is greater than 75% (ie the median of the distribution of Spanish banks in 2014). For the reason that literature suggests a number of channels via which damaging rates of interest have an effect on banks (ie retail deposits, extra liquidity, floating-rate loans, short-term interbank positions), the BLS gives a abstract measure of publicity to them.
We derive the affected dummy from a probit regression the place our dependent variable is a dummy which information if the financial institution reported that the ECB’s damaging deposit facility price contributed to a lower of the financial institution’s internet curiosity earnings (NII) previously six months, and the regressors embrace the deposit ratio and the liquidity ratio. As well as, affected banks might have a excessive share of floating-rate loans or short-term loans, that are repriced at a decrease price following a discount within the official rate of interest. Subsequently, we additionally embrace the load of mortgage overdrafts and loans with a maturity as much as one yr within the complete inventory of loans, respectively, and management for different financial institution traits corresponding to solvency, profitability, dimension and Eurosystem borrowing.
Within the second step, we then run regressions utilizing a 3 dimensional (bank-borrower-time) panel, regressing financial institution lending on the therapy variable, firm-time and financial institution particular mounted results.
Banks’ self-assessment of the impression of damaging rates of interest on their steadiness sheets might pose an identification problem, as weak banks might have incentives to strategically misreport their analysis of the coverage in an effort to ‘blame’ it for his or her poor efficiency. Nonetheless, we acquire comparable ends in a number of robustness analyses solely primarily based on arduous information, wherein we classify banks in keeping with their deposit ratios or their share of credit score at floating charges.
Importantly, we permit for various results in several intervals by interacting our key regressor with time dummies, to investigate the dynamic impression of damaging rates of interest between 2014 and 2019, a interval wherein deposit charges in Spain exhibited a downward pattern till reaching the ZLB. We additionally tackle two key identification challenges. First, we disentangle credit score provide from credit score demand by together with firm-time mounted results à la Khwaja and Mian (2008), which means evaluating lending selections of a number of banks to the identical agency throughout the similar interval. Second, we management for the potential results of different ECB insurance policies on banks’ credit score provide, such because the focused long-term refinancing operations (TLTROs) and the expanded asset buy programme (APP).
4 key outcomes
First, we discover that affected banks contracted their lending provide to companies solely over the past sub-sample interval (2018–19), whereas there isn’t any impact in the course of the earlier intervals. This end result could also be defined by the truth that retail deposit charges have been excessive in Spain on the time of the introduction of the damaging rates of interest, so they’d loads of room to say no earlier than reaching the ZLB in 2018. Nonetheless, since coverage charges within the euro space have been lowered a number of instances since 2014, we can’t rule out a complementary rationalization, specifically that coverage charges reached the reversal price (Brunnermeier and Koby (2019)), which is the speed at which accommodative financial coverage ‘reverses’ its meant impact and turns into contractionary for lending.
Second, we additionally discover that the impact of damaging rates of interest on banks’ credit score provide was heterogeneous and relied on the extent of banks’ capitalisation. Particularly, we observe that affected banks with low capital ratios curtailed their lending provide to companies, however they solely did so over the past interval 2018–19, when deposit charges reached the ZLB. Therefore, our findings recommend that the reversal price could be bank-specific and depending on banks’ capitalization ranges.
Third, splitting our pattern into secure and dangerous companies, we discover that affected low-capitalised banks lowered their credit score provide to dangerous companies within the final two pattern intervals, 2016–18 and 2018–19, though the impact is far stronger within the latter interval. Against this, there’s solely a touch vital impact on secure companies within the final interval, and its dimension is considerably smaller than that for dangerous companies. Subsequently, our findings point out that affected low-capitalised banks contracted their credit score provide to dangerous companies previous to limiting it to secure companies and in a larger magnitude, arguably as a result of loans to the previous devour extra regulatory capital than exposures to the latter. This proof means that affected low-capitalised banks took much less threat due to their lack of capital buffers to soak up losses and the necessity to meet capital necessities. Fourth, trying on the impression of the damaging rates of interest on companies’ complete borrowing we discover no proof that corporations whose foremost credit score establishment was an affected low-capitalised financial institution skilled a contraction of their complete financial institution credit score. This proof means that the decrease provide of credit score by affected low-capitalised banks was offset by the upper lending provide by non-affected banks, with capability for taking further dangers due to their greater capital buffers. Subsequently, whereas the reversal price could be reached by some affected undercapitalised banks, there gave the impression to be no combination impact on the provision of lending to companies.
Miguel García-Posada and Sergio Mayordomo work at Banco de España.
If you wish to get in contact, please e mail us at bankunderground@bankofengland.co.uk or depart a remark under.
Feedback will solely seem as soon as authorized by a moderator, and are solely printed the place a full identify is provided. Financial institution Underground is a weblog for Financial institution of England employees to share views that problem – or assist – prevailing coverage orthodoxies. The views expressed listed here are these of the authors, and are usually not essentially these of the Financial institution of England, or its coverage committees.
