Monetary Policy and Transmission Mechanism in Thailand by Dr. Atchana Waiquamdee Deputy Governor, Bank of Thailand February 2008
Presentation Outline 2 Monetary Transmission Mechanism Interest Rates and Bank Lending Exchange Rate Asset Prices Expectations Relative importance of key transmission channels Major changes influencing MP transmission since the 1997 financial crisis Banking disintermediation Excess liquidity in the Thai banking system in 2003-2005
Monetary Policy Transmission 3 Transmission channels Central bank S-T Interest Rate Interest rates Bank lending Exchange rate Output and OMO Asset prices Inflation Expectation
Pass-through via interest rates and bank lending 4 Policy rate to money market rates and the term structure of interest rates Policy rate to commercial bank rates (Deposit & MLR) Dynamic Multiplier Model (DMM)
Money market rates and short-term bond yields respond quickly to and in the same direction as the current policy rate change 5 % 5.0 Money Market Rates % 8 Government Bond Yields 4.5 7 14 Y 4.0 3.5 3.0 6 5 4 10 Y 5 Y 3 Y 2 Y 1 Y 2.5 2.0 1.5 14D R/P 3 2 3 M 1.0 Interbank 1D R/P 1 0.5 Jan 03 Jul-03 Jan 04 Jul 04 Jan 05 Jul 05 Jan 06 0 2-Jun-03 28-Aug-03 21-Nov-03 19-Feb-04 24-May-04 20-Aug-04 15-Nov-04 11-Feb-05 17-May-05 15-Aug-05 8-Nov-05 3-Feb-06 9-May-06
Long-term bond yields depend on market expectations of future policy rate changes, but the response is also quick 6 % p.a. 7 Thai Government Bond Yield Curve 6 5 4 3 2 1 29 Jun 2006 30 Dec 2005 30 Dec 2004 31 Aug 2004 5 Jan 2004 0 1m 3m 6m 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 TTM (years)
GDP growth response to an interest rate shock according to BOTMM 7 Interest Rate Channel As of Jan 2003 GDP %deviation from Baseline - - -0.75-1.00-1.25-1.50 All Channel -1.75 W/O Interest Rate -2.00 Q1 Q5 Q9 As of Jul 2006 GDP %deviation from Baseline - - -0.75-1.00-1.25-1.50 All Channel -1.75 W/O Interest Rate -2.00 Q1 Q5 Q9 The magnification of transmission via interest rate channel could be explained by the adoption of Inflation Targeting that uses interest rate as the policy signal and the increased importance of bond market
Pass-through via interest rates and bank lending 8 Policy rate to money market rates and the term structure of interest rates Policy rate to commercial bank rates (Deposit & MLR) Dynamic Multiplier Model (DMM)
Policy retail interest rates 9 % p.a. 30 25 Fixed Exchange Rate Regime Monetary Targeting Floating Exchange Rate Regime Inflation Targeting (RP rate as policy instrument) 20 15 RP14 3m deposit MLR 10 5 0 Jan-90 Jan-92 Jan-94 Jan-96 Jan-98 Jan-00 Jan-02 Jan-04 Jan-06 Source: Bank of Thailand
Pass-through* from policy rate to retail rates 10 1.2 The degree of interest rate pass-through changes over time, depending to a significant extent on how well the banking sector s intermediary role functions 1.0 0.8 0.6 Long run impact (MLR) Around 2003-2005, the weakened pass-through was partly due to excess liquidity in the banking system 0.4 0.2 0.0 1993M01 1994M01 1995M01 1996M01 1997M01 1998M01 1999M01 2000M01 2001M01 2002M01 2003M01 2004M01 2005M01 2006M01 2007M01 * Using the dynamic multiplier method, i.e., regressing DMLR on DRP14d (contemporaneous and lagged) and lagged DMLR with a rolling window of 50 observations
Pass-through from policy rate to retail rates 11 Degree of pass-through of policy rate to 3-mth deposit rate (RD3M) and MLR as estimated by the dynamic multiplier model Interest rates Period Impact Immediate 3-month 6-month Long-run RD3M 1989m1 1995m12 0.072 0.427 0.608 0.744 2000m1 2007m12 0.376 0.637 0.679 0.725 MLR 1989m1 1995m12 0.090 0.327 0.520 0.557 2000m1 2007m12 0.270 0.409 6 0.554 Despite a temporary period of impaired transmission, there is evidence that the pass-though from policy rate to retail rates has picked up in the most recent years. There is also evidence that this transmission channel has become quicker (but not stronger as a whole) since 2000 compared to the pre-crisis period.
International comparison of MLR pass-through 12 Country Impact 3 months 6 months Long Run Germany 0.38 0.67 0.83 1.04 Australia -- 0.35 0.67 0.81 US 0.41 0.97 0.97 0.97 UK 0.82 1.02 1.04 1.04 Denmark 0.07 0.38 0.71 Finland 0.13 0.2 0.27 0.6 Singapore 0.27 0.71 0.82 0.95 Indonesia 0.2 0.74 0.74 1.00 Philippines 0.24 0.64 0.64 0.64 Malaysia 0.13 0.28 0.37 0.44 Japan 0.03 0.22 0.35 0.53 Thailand 0.27 0.41 0.51 0.55
GDP growth response to an interest rate shock according to BOTMM 13 Bank Lending Channel As of Jan 2003 As of Jul 2006 - - -0.75-1.00 GDP %deviation from Baseline - - -0.75-1.00 GDP %deviation from Baseline -1.25-1.25-1.50 All Channel -1.50 All Channel -1.75 W/O Banking -1.75 W/O Banking -2.00-2.00 Q1 Q5 Q9 Q1 Q5 Q9 Further discussion on bank lending channel in the following section Banking disintermediation Excess liquidity
Interest Rate and Bank Lending Channels 14 Transmission from the policy rate to money market and bond yields, both short- and long-term, functions quite well. With the increased importance of the bond market, this channel has gained importance over the years. The pass-through to commercial banks rates and lending, however, was temporarily impaired after the financial crisis, though it has returned to normal in the most recent years.
Exchange Rate Channel remains important 15 According to econometric results, the interest rate differential is found to have a small but statistically significant effect on Thailand s exchange rate movements* Nevertheless, the impact of exchange rate on the real economy is always significant. Effect of 1% depreciation in the nominal effective exchange rate (NEER)** Economic growth Headline inflation Core inflation As of Jul 06 +0.26 +0.06 +0.06 Note: **average effect in 1 year Note:* A 1% increase in the interest rate differential will cause the THB exchange rate to appreciate by 0.088%, D(FX) = -1.996e-06*D(BP(-1)) +0.153*D(YENDOLLAR) -0.088*INTDIFF +[AR(1)=0.426] GARCH = 3-0.060*RESID(-1)^2 +1.027*GARCH(-1)
GDP growth response to an interest rate shock according to BOTMM 16 Exchange Rate Channel As of Jan 2003 As of Jul 2006 GDP %deviation from Baseline - - -0.75-1.00-1.25-1.50 All Channel -1.75 W/O FX -2.00 Q1 Q5 Q9 GDP %deviation from Baseline - - -0.75-1.00-1.25-1.50 All Channel -1.75 W/O FX -2.00 Q1 Q5 Q9
Asset Price Movements: House and Stock Prices 17 Index (Year 2000 = 100) 130 Index (30 Apr 1975 = 100) 1600 120 110 Single-detached house 1400 1200 1000 100 800 90 Town house SET (RHS) 600 400 80 200 70 0 Q1 1993 Q1 1994 Q1 1995 Q1 1996 Q1 1997 Q1 1998 Q1 1999 Q1 2000 Q1 2001 Q1 2002 Q1 2003 Q1 2004 Q1 2005 Q1 2006 Source: Government Housing Bank, SET and Bank of Thailand
Asset Price Channel remains weak in Thailand 18 Most of household wealth is in the form of residential and commercial real estate (non-financial assets). These assets are quite illiquid, while equity withdrawal from these assets is still limited in Thailand. Even financial assets held by households are mostly deposit accounts, not securities whose values are sensitive to changes in the policy interest rate Composition of Household wealth Composition of Financial Assets 2% 29% 43% 13% 13% 23% 15% 5% 55% 2% Vehicles RealEstate (residential) RealEstate (commercial) Financial Assets Livestock Deposit accounts Providence funds, insurance Others Stocks, bonds, mutual funds Cash and jewellry Source: National Statistical Office s Socio-Economic Survey, 2006
GDP growth response to an interest rate shock according to BOTMM 19 Asset Price Channel As of Jan 2003 As of Jul 2006 GDP %deviation from Baseline - - -0.75-1.00-1.25-1.50 All Channel -1.75 W/O Asset -2.00 Q1 Q5 Q9 GDP %deviation from Baseline - - -0.75-1.00-1.25-1.50 All Channel -1.75 W/O Asset -2.00 Q1 Q5 Q9
Comparison Across Transmission Channels as of January 2003 20 GDP %deviation from Baseline - - -0.75-1.00-1.25-1.50 All Channel -1.75 W/O Asset -2.00 Q1 Q5 Q9 GDP %deviation from Baseline - - -0.75-1.00-1.25-1.50 All Channel -1.75 W/O Banking -2.00 Q1 Q5 Q9 GDP %deviation from Baseline - - -0.75-1.00-1.25-1.50 All Channel -1.75 W/O FX -2.00 Q1 Q5 Q9 GDP %deviation from Baseline - - -0.75-1.00-1.25-1.50 All Channel -1.75 W/O Interest Rate -2.00 Q1 Q5 Q9
Comparison Across Transmission Channels as of July 2006 21 GDP %deviation from Baseline - - -0.75-1.00-1.25-1.50 All Channel -1.75 W/O Asset -2.00 Q1 Q5 Q9 GDP %deviation from Baseline - - -0.75-1.00-1.25-1.50 All Channel -1.75 W/O Banking -2.00 Q1 Q5 Q9 GDP %deviation from Baseline - - -0.75-1.00-1.25-1.50 All Channel -1.75 W/O FX -2.00 Q1 Q5 Q9 GDP %deviation from Baseline - - -0.75-1.00-1.25-1.50 All Channel -1.75 W/O Interest Rate -2.00 Q1 Q5 Q9
Monetary policy transmission in Thailand 22 Monetary policy transmission is primarily through the interest rate and exchange rate channels However, the sensitivity of retail interest rates to the policy interest rate was temporarily impaired after the financial crisis Asset price channel has yet to develop
Factors influencing the transmission channels 23 Banking disintermediation Excess liquidity in Thai banking system during 2003-2005
Firms reliance on bank financing declined as banks balance sheets deteriorated and risk aversion rose significantly in 1997 24 Financing Relative to GDP (1960-2005) Percent of GDP 1997 crisis 160% 140% 120% 100% Commercial banks SFIs Stocks Finance companies Bonds 80% 60% 40% 20% 0% 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 Note: Commercial banks, finance companies and credit foncier companies and SFIs (total assets); stocks (SET market capitalization); bonds (outstanding values of public and corporate bonds at par value) Sources: BOT; SET; TBDC
A more limited role of bank lending 25 % 140 120 100 The ratio of bank credits to GDP has fallen since the financial crisis, confirming a significant degree of disintermediation. The recovery in GDP has not been accompanied by a comparable bank credit expansion. Bank credit / GDP 25 20 15 10 GDP vs. Real credit growth 80 5 60 0 40-5 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 20-10 0-15 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007-20 Source: BOT, NESDB
A more limited role of bank lending 26 Bond and equity markets have become important alternative sources of financing for large firms. Billion Baht 600 Private Sector s Sources of Funds 400 200 0-200 -400-600 -800 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 Bank loans (Change in stock) SFI loans (Change in stock) Stocks (New issue) Bonds (New issue excl short-term bills)
Factors influencing the transmission channels 27 Banking disintermediation Excess liquidity in Thai banking system during 2003-2005
Transmission through R/P on the asset side of a bank s balance sheet is less effective 28 Bond Loan Deposit R/P R/P Bond Loan Deposit Returns Loans MLR = Cost of funding + spread Deposits R/P R/P Unusually low rate Risks Billion baht 500 400 300 200 100 0-100 Net R/P Lending of Banks 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005
Conceptual Framework 29 Based on specifying the proportion of disposable liquidity as excess Disposable liquidity includes public-sector securities, net foreign asset (NFA), and net lending in R/P Billion baht 700 600 500 400 300 200 100 0 Disposable Liquidity Securities NFA NL in RP Jul-04 Oct-04 Jan-05 Apr-05 Jul-05 Oct-05 Jan-06 Source: BOT calculation
Conceptual Framework (cont.) 30 The dynamic process is based on the relative risk-return profile of each asset For example, an asset is considered excess if the risk-adjusted returns are low relative other assets such as loans. However, as its relative returns rise, banks will be more willing to hold, and it will become less excess. excess
Relative Returns: returns on liquid assets have risen higher compared with deposit and lending rates 31 Relative Returns of Liquid Assets to Loan 1.00 0.75 Gov/MLR LIBOR/MLR RP/MLR % 5 4 3 2 Deposit rates of 4 large banks 14-D RP 12 M 5.00 4.00 3. 50 3. 25 1 6 M 3 M Jul-04 Jan-05 Jul-05 Jan-06 0 May-04 Nov-04 May-05 Nov-05 May-06 resulting in higher demand for liquidity (i.e. more willing to hold liquid assets like R/P)
Thus, excess liquidity decreased from around 15% of deposits at the beginning of the tightening cycle to around 6% of deposit 32 Billion baht % of deposits 800 Measure of Excess Liquidity 20% 600 16% 400 200 386 12% 8% 5. 90% 4% 0 Jul-04 Oct-04 Jan-05 Apr-05 Jul-05 Oct-05 Jan-06 Apr-06 0% Securities** Net RP** NFA** EL/Deposits Source: BOT calculation