Banque nationale de Belgique SA · EURONEXT:BNB
Banque nationale de Belgique SA — Technical Presentation (June 2015)
40 slides from Banque nationale de Belgique SA's technical presentation, Jun 22, 2015
Slide 1: The Relationship Between Bank Lending Rates, Policy Rates and Bank Funding Costs After the Global Financial Crisis
The Relationship Between Bank Lending Rates, Policy Rates and Bank Funding Costs After the Global Financial Crisis by Anamaria Illes, Marco J Lombardi and Paul Mizen National Bank of Belgium, 22 June 2015 - The views expressed are solely those of the author and should not be attributed to the BIS Re
Slide 2: Key Issues
Key Issues The global financial crisis prompted central banks in many countries to cut short-term policy rates to near zero levels after the Lehman collapse in September 2008 Based on the pre-crisis relationship between bank lending rates and policy rates, it would have been reasonable to expect
Slide 3: Short term lending and policy rates in the UK
Short term lending and policy rates in the UK Restricted 3
Slide 4: Short term lending and policy rates in the EA
Short term lending and policy rates in the EA Restricted 4
Slide 5: Policy rates and bank lending rates
Policy rates and bank lending rates Three reasons why they do not follow policy rates First, the policy rate is a very short-term rate, and so ignores the maturity mismatch with lending rates. Second, even if we correct for the maturity mismatch using an appropriately adjusted swap rate, the adj
Slide 6: Policy rates and bank lending rates
Policy rates and bank lending rates Reasons why policy rates are less relevant after the crisis Since the global financial crisis there have been a number of changes that have increased the cost of bank funding Larger risk premia on securities issued by banks and interbank borrowing, have raised t
Slide 7: Policy rates and bank lending rates
Policy rates and bank lending rates Why this has become a major policy issue Central Banks are concerned about the possible ineffectiveness of monetary policy. Central banks want to understand how unconventional policy affects bank lending rates and volumes. The transmission of monetary policy
Slide 8: Funding costs and bank lending rates
Funding costs and bank lending rates We construct a weighted average cost of liabilities for banks, which reflects the increase in the cost of funds that they have experienced. It comprises a volume-weighted average of the rates at which banks can obtain finance: WACLit j 1 wijt rijt J where ri
Slide 9: Funding costs and bank lending rates
Funding costs and bank lending rates A key issue in the paper is therefore the construction of the weights in the WACL measure of bank funding costs. The WACL is constructed using five types of liabilities - retail deposits, senior unsecured or covered bonds and the interbank market, central bank op
Slide 10: Component weights
Component weights deposit liabilities (in all currencies, and excluding the general government) vis-à-vis the euro area to MFIs and to non-MFIs. Deposit liabilities to MFIs are equivalent to interbank deposits, while deposits to non-MFIs correspond to deposits of the private non-financial sector
Slide 11: Component weights
Component weights debt securities, issued in all currencies in the euro area by the MFIs. This is obtained from the MFIs balance sheets, which also report breakdowns of maturities up to and over one year. Bond markets are segmented to a large extent on national lines (van Rixtel and Gasperini, 2
Slide 12: Component weights
Component weights covered bonds data are from Dealogic. This component is only used for the long-term calculations since all of the covered bonds have a maturity of more than one year. The volume of covered bonds outstanding is not large, but has grown since the crisis. Banks in Austria, Finland
Slide 13: Component weights
Component weights central bank operations for the euro area countries only becomes important after 2009. Van Rixtel and Gasperini (2013) show that Spain and Italy relied most heavily on liquidity operations provided by the ECB having borrowed, respectively, €400bn and €277bn in September 2012.
Slide 14: Component weights
Component weights equity issuance is excluded. It accounts for a small percentage of the outstanding balances, and it is arguably not used by banks as a source of regular finance for bank lending, but rather as a structural adjustment (e.g. adjustment of capital ratios in response to regulatory
Slide 15: Weights of the short term WACL
Weights of the short term WACL Restricted 15
Slide 16: Weights of the long term WACL
Weights of the long term WACL Restricted 16
Slide 17: Funding costs and bank lending rates
Funding costs and bank lending rates Adjusted stocks-based WACL. We assume banks only use lowest cost funding options. Sharp increase in credit spreads on conventional bonds issued by banks can raise these rates above lending rates therefore we allocate greater weight to the covered bonds category
Slide 18: Funding costs and bank lending rates
Funding costs and bank lending rates Interest rates used are always based on new transactions data. For MFIs deposits we use the overnight and 1-year interbank money market rates for ST and LT. For non-MFI deposits we use ECB MFI interest rate statistics. For debt securities, we take the OIS
Slide 19: Selected WACL funding costs and policy rates for France, Germany and the UK
Selected WACL funding costs and policy rates for France, Germany and the UK France Germany UK Restricted 19
Slide 20: Selected WACL funding costs and policy rates for Ireland, Italy, Portugal and Spain
Selected WACL funding costs and policy rates for Ireland, Italy, Portugal and Spain Ireland Italy Portugal Spain Restricted 20
Slide 21: Estimation by pooled mean group (PMG) estimator
Estimation by pooled mean group (PMG) estimator We take yit, (the lending rate), and xit, (the driver of the lending rate i.e. policy rate or the WACL). Assuming an ARDL model yit p 1 ip yit p q 0 iq xit q i eit P Q we rewrite as a stacked set of N individual equations relating
Slide 22: Estimation by pooled mean group (PMG) estimator
Estimation by pooled mean group (PMG) estimator After establishing the order of integration using Im, Pesaran and Shin tests, the model has the following specification: DYi i (Yi , 1 q i X i ) p 1 ip DYi , p q 0 iq DX i , q i 1 i P 1 Q 1 where the relationship i (Y
Slide 23: Results – Main messages
Results – Main messages Data are non-stationary according to Im-Pesaran-Shin tests. Cointegration relationships exist for the full sample for the WACL(stock) and WACL(flow) v.s lending rates with a few exceptions. NO cointegration exists for policy rates v.s lending rates on the full sample. P
Slide 24: Unit Root Tests
Unit Root Tests Im-Pesaran-Shin tests Restricted 24
Slide 25: Cointegration tests
Cointegration tests Restricted 25
Slide 26: Cointegration tests
Cointegration tests Restricted 26
Slide 27: Results – Main messages (continued)
Results – Main messages (continued) Results for WACL(flows) are not statistically different from those reported for WACL(stocks) despite different weights. Taking all 11 countries, EA and EA(core), PMG estimates imply short-term lending rates and WACL(stocks) relationships are fairly similar a
Slide 28: PMG estimates of pass through – WACL (stock)
PMG estimates of pass through – WACL (stock) Restricted 28
Slide 29: PMG estimates of pass through – WACL (flow)
PMG estimates of pass through – WACL (flow) Restricted 29
Slide 30: Results – contrast with policy rates
Results – contrast with policy rates Pass-through estimates with policy rates: 100bp ↑ in policy rates 52-96bp ↑ in lending rates. The range of values is wider. In many cases the pass through estimate is lower. There is a structural break in policy rates and cointegration breaks down even when
Slide 31: PMG estimates – policy rates
PMG estimates – policy rates Restricted 31
Slide 32: Results – robustness
Results – robustness We consider sub-samples to explore the pass-through of WACL to lending rates up to three alternative break points: 1) the global financial crisis (GFC), July 2007, after which bank funding costs increased significantly. 2) the Lehman collapse and subsequent recession 3) the sove
Slide 33: Sub sample estimation using WACL (stock)
Sub sample estimation using WACL (stock) Restricted 33
Slide 34: Impulse responses using NFC short v WACL (stock)
Impulse responses using NFC short v WACL (stock) Restricted 34
Slide 35: Impulse responses using NFC long v WACL (stock)
Impulse responses using NFC long v WACL (stock) Restricted 35
Slide 36: Impulse responses for mortgage short v WACL (stock)
Impulse responses for mortgage short v WACL (stock) Restricted 36
Slide 37: Impulse responses for mortgage long v WACL (stock)
Impulse responses for mortgage long v WACL (stock) Restricted 37
Slide 38: Implications
Implications Heterogeneity in financial conditions across the euro area has increased significantly, as some countries have been affected more substantially by the financial crisis. Money markets have become impaired, especially across national borders, and yields in sovereign bond markets have
Slide 39: Implications
Implications For these reasons … (the bullets on the previous slide are all verbatim quotes from ECB monthly bulletins Aug 2012 pp 63-4, 68 and Aug 2013 p75) …financing conditions have more to do with financial conditions in different countries -- which are heterogeneous in the post crisis period –
Slide 40: Conclusions
Conclusions Policy rates have fallen sharply while lending rates have not. Some conclude that banks have failed to pass through rate changes. We argue that funding costs are disconnected from policy rates in the post crisis period. Some funding costs have risen with elevated credit risk. The L
Banque Nationale De Belgique — Technical Presentation (June 2015) on 6ix