Transcript Optimal Regulation Of Bank Capital And Liquidity

OPTIMAL REGULATION OF
BANK CAPITAL AND
LIQUIDITY
Course on Financial Instability at the Estonian Central Bank,
9-11 December 2009 – Lecture 8
E Philip Davis
NIESR and Brunel University
West London
[email protected]
www.ephilipdavis.com
groups.yahoo.com/group/financial_stability
Abstract
• Raising capital adequacy standards and introducing
binding liquidity requirements can have beneficial
effects if they reduce the probability of a costly
financial crisis, but may also reduce GDP by raising
borrowing costs for households and companies.
• We estimate both benefits and costs of raising capital
and liquidity, with the benefits being in terms of
reduction in the probability of banking crises, while
the costs are defined in terms of the economic impact
of higher spreads for bank customers.
• Result shows a positive net benefit from substantial
regulatory tightening, depending on underlying
assumptions.
Structure
1.
2.
3.
4.
5.
6.
7.
Introduction
Literature survey
Could we predict the crisis?
Impact of the crisis
Modelling the banking sector
Costs and benefits of tighter regulation
Conclusion
1 Introduction
• NIESR research for FSA - background is recent
crisis and discussion of regulatory changes at
global level – authors Ray Barrell, E Philip Davis,
Tatiana Fic, Dawn Holland, Simon Kirby and Iana
Liadze
• Review literature on bank behaviour relative to
regulation of capital and liquidity
• Then assess whether a link of capital and liquidity
to crisis probabilities can be traced
• Consider long run cost of crises (scarring)
• Model behaviour of banks, impact on economy
• Consider over what range there are positive net
benefits to regulatory tightening
2
Literature survey
• Much theoretical work abstracts from regulation
• Most relevant is on bank capital buffers…
• …and UK empirical work showing regulation
(trigger ratios) is main determinant of capital held –
and also impacts loan supply
• Overall procyclicality of capital and balance sheet,
and impact of introduction of Basel in early 1990s
(credit crunch)
• Cross country work showing impact of capitalisation
on margins
• Controversy regarding relevance of MM
• Developing literature on crisis probabilities
trigger ratio
risk weighted capital ratio
Dec-07
Dec-06
Dec-05
Dec-04
Dec-03
Dec-02
Dec-01
Dec-00
Dec-99
Dec-98
Dec-97
Dec-96
Dec-95
Dec-94
Dec-93
Dec-92
Dec-91
Dec-90
Dec-89
Percent of total assets
UK bank capital adequacy
14.00
13.00
12.00
11.00
10.00
9.00
8.00
7.00
6.00
3
Predicting crises
• Most work on predicting crises, such as Demirguc
Kunt and Detragiache (1998) uses logit and
estimates across global sample of crises (mainly in
emerging market economies)
• Typically crises found to be correlated with
macroeconomic, banking sector and institutional
indicators:
– Low GDP growth, high interest rates, high inflation,
fiscal deficits.
– Ratio of broad money to Foreign Exchange reserves,
credit to the private sector/GDP ratio, lagged credit
growth
– Low GDP per capita and deposit insurance.
Could they predict the crisis?
• Traditional crisis prediction models did not pick
up the risks that were developing (Davis and
Karim 2008)
–
–
–
–
The build up of debt was worrying
The house price bubble was a concern
The regulatory architecture was flawed
The dangers of securitisation were not seen
• None of these were under the control of the
monetary authorities
• A crisis means credit rationing
(Barrell et al 2006 Journal of Financial Stability)
Explaining Crises
Probabilities of a crisis
After 1
After 2
After 3
Initial
percentage
percentage
percentage
probability
point increase point increase point increase
of the cirsis
in LIQ and
in LIQ and
in LIQ and
in 2007
LEV
LEV
LEV
FR
GE
IT
SP
UK
US
0.066
0.007
0.035
0.056
0.217
0.006
0.043
0.005
0.023
0.024
0.150
0.004
0.028
0.003
0.014
0.023
0.101
0.003
0.018
0.002
0.009
0.015
0.067
0.002
Variables are tak en with lag length (based on the model)
 p(crisis) 
log 
 = - 0.333 LEV(-1) – 0.118 LIQ(-1) + 0.113 RHPG(-3)
 1 - p(crisis) 
(-2.85)
(-3.55)
(2.8)
(3)
Comments on results
• We use unadjusted capital adequacy due to data
limitations over 1980-2006 sample
• Changes in capital alone twice as effective as
liquidity according to estimate
• Results show substantial reduction in crisis
probabilities from regulatory tightening in Europe
using OECD logit
• Weaker effect in US, which we consider relates to
omission of off balance sheet activity – work
under way to rectify this
4
Impact of the crisis
• Seek to assess effect of current crisis on UK
economy, as a benchmark for benefits of
regulation
• Key background is approach of Hoggarth and
Sapporta (2001) who saw costs of crisis as integral
of output lost below previous trend
• Highlight that initial recession is only part of costs
given the possible long run effect on capital stock
(scarring)
Assessing long run costs of crises
Output depends on the supply side as described by the production function
Qt = γ (δKt-ρ + (1-δ)(
techl -ρ -1/ρ
e
t
t) )
L
The equilibrium capital output ratio depends on the user cost of capital
Log (Kt /Qt) = a1 + σ
L
o
g
(
u
s
e
r
t)
The user cost is driven by weighted average cost of capital, linked in
turn to risk free long real rates (lrr) and by the borrowing margin
charged by banks (corpw) or the bond market (iprem) to reflect costs
and risks
wacct = b1t(Et/Pt) +(1-b1t) (c1t (lrrt+corpwt)
+(1- c1t) (lrrt+ipremt))*(1-ctaxrt)
US
Euro Area
UK
05 June 2009
05 March 2009
05 December 2008
05 September 2008
05 June 2008
05 March 2008
05 December 2007
05 September 2007
05 June 2007
05 March 2007
05 December 2006
05 September 2006
05 June 2006
05 March 2006
05 December 2005
05 September 2005
05 June 2005
05 March 2005
05 December 2004
05 September 2004
05 June 2004
05 March 2004
05 December 2003
05 September 2003
05 June 2003
05 March 2003
05 December 2002
05 September 2002
05 June 2002
05 March 2002
05 December 2001
05 September 2001
05 June 2001
05 March 2001
05 December 2000
05 September 2000
05 June 2000
05 March 2000
Risk premia on corporate borrowing
rose in 2007-9
Spread between BAA corporate and government bonds
10
9
8
7
6
5
4
3
2
1
0
Trend UK output and the scar
• Trend output has stepped
down
12.9
12.8
July 2008 forecast
12.7
12.6
12.5
95, 90 and 80% confidence bounds
around April forecast
2018
2016
2014
2012
2010
2008
2006
2004
2002
12.4
2000
• There is only a 1 in 20
chance output will be at
July 2008 projections in
2018
13
Output (log scale)
– Risk premia rose in two
stages before and after
Lehman’s –this reduced
trend 3-4%
– Banking sector gains
were illusary (1-2%)
– In-migration will fall
reducing trend by ¾%
Output projections and scarring
from the Crisis
5
Modelling the banking sector
• Shown that regulation has benefits in reducing
incidence and costs of crises
• May also have a negative impact on output in both
the short and the long run by increasing borrowing
costs and raising user cost of capital (an effective
tax on banks has real effect, e.g. widens spreads)
• So model effects of regulation on output by
constructing banking sector model and embedding
in NiGEM
• Use the risk weighted capital adequacy, but
correlation of 0.92 to unweighted
NiGEM model
• NiGEM covers the OECD economies
–
–
–
–
Supply and demand spelled out
Trade and capital account linkages
Stock flow consistent
Long run properties as a DSGE model
• Financial and exchange markets forward looking, as is the
wage bargain
• Capital stock depends on user cost and on expected output 4
years ahead
• Increasing spread between borrowing and lending rates for
individuals changes their incomes, and decision making on the
timing of consumption, with possibility of inducing sharp
short term reductions (no long run effect).
• Changing spread between borrowing and lending rates for
firms may change user cost of capital and hence equilibrium
level of output and capital in the economy in a sustained
manner (short and long run effects).
UK banking sector model
• Banking activity modelled as a set of supply (or price) and
demand curves.
• Demand depends on levels of income or activity, and on
relative prices, whilst supply, or price, depends upon the
costs of providing assets and on the risks associated with
those assets.
• Banking sector has four assets
– secured loans to individuals for mortgages, (morth) with a
borrowing cost (rmorth),
– unsecured loans to individuals for consumer credit (cc) with a
higher borrowing cost or rate of return (ccrate),
– loans to corporates (corpl) with a rate or return or cost of
borrowing (lrr+corpw) where lrr is the risk free long rate and
corpw is the mark up applied by banks
– liquid assets (lar).
– The categories subsume, along with deposits and risk weighted
capital adequacy itself (levrr), all on-balance sheet activity within
the UK.
Key equations
• Corporate spread related to capital
adequacy and inverse of headroom to
trigger ratio
corpw = -0.196809 + 0.131227*(log(y)-log(ycap))*100
(-0.46)
(4.69)
+ 0.841752*invhead(-1) + 0.522302*insolr + 0.194533*levrr
(5.32)
(5.72)
(5.60)
• Household spread related to capital
adequacy
lendw = lendw(-1) - 0.000128 - 0.446002*(lendw(-1) +0.024201
(-0.87)
(-4.41)
(5.83)
- 0.063713*levrr(-3)
(2.84)
- 0.006035*arr(-4) + 0.003159*nwpi(-1))
(8.75)
(-7.28)
Model details
• Each spread feeds into borrowing cost and then
credit volume, with appropriate signs
• No direct role for liquidity (effect is calibrated
based on estimated results for US)
• One to one relation of bank spread to bond spread
(arbitrage)
• Adjustment equations for balance sheet and capital
Absolute change in corporate borrowing costs
Tighter regulation and margins
1.8
1.6
1.4
1.2
1
0.8
0.6
0.4
0.2
0
2009Q1
2010Q1
2011Q1
2012Q1
2013Q1
2014Q1
One percentage point rise in capital
Three percentage point rise in capital
2015Q1
2016Q1
2017Q1
2018Q1
Two percentage point rise in capital
Note – changes to target capital only
Tighter regulation and output
0.05
Percent change in GDP
0
2009Q1
2010Q1
2011Q1
2012Q1
2013Q1
2014Q1
2015Q1
2016Q1
2017Q1
2018Q1
-0.05
-0.1
-0.15
-0.2
-0.25
One percentage point rise in capital
Three percentage point rise in capital
Two percentage point rise in capital
Note – equal changes to target and actual capital in this and following
(6) Costs and benefits of tighter
regulation
• Compare the gains from tighter regulation to
the costs of regulation
– Gains are the change in the probability of a crisis
times lost output in this crisis
– Costs are regulatory impacts
• Look at steady state impacts in 2018
Borrowing costs
Impacts on borrowing costs of increases in regulatory constraints
1
2
5
10
Changes in Capital and liquidity Requirements
Corporate borrowing costs (absolute change)
One percentage
Two Percentage Three percentage
point increase
point increase
point increase
0.29
0.59
0.88
0.29
0.58
0.87
0.29
0.58
0.87
0.29
0.58
0.87
1
2
5
10
User cost of capital (percent change)
0.63
1.26
0.77
1.54
0.85
1.70
0.85
1.70
1.88
2.28
2.52
2.50
2.50
3.02
3.33
3.30
1
2
5
10
Changes in Liquidity Requirements
Corporate borrowing costs (absolute change)
0.10
0.19
0.10
0.20
0.11
0.22
0.11
0.23
0.29
0.30
0.33
0.34
0.39
0.40
0.43
0.45
1
2
5
10
User cost of capital (percent change)
0.21
0.42
0.27
0.53
0.32
0.64
0.33
0.67
0.63
0.79
0.93
0.97
0.84
1.05
1.23
1.27
Year
Four percentage
point increase
1.17
1.17
1.16
1.16
Long run - regulation on output
2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026
0.1
0
percentage points of GDP
-0.1
-0.2
-0.3
-0.4
-0.5
-0.6
-0.7
-0.8
-0.9
one
two
three
four
five
six
seven
Cost benefit analysis - combined
0.08
net present value as a proportion of GDP
0.07
0.06
0.05
0.04
0.03
0.02
0.01
0
-0.01
-0.02
-0.03
1
2
3
4
5
6
Percentage points increase in capital adequacy and liquidity standards
7
7
Conclusions
• Lax regulation raises the risk of a crisis
• Raising capital adequacy standards and
introducing binding liquidity requirements
beneficial if they reduce probability of costly
financial crisis
• Also costs - any effective banking regulation
works as a tax on bank activity. Hence regulations
may reduce output through impacts on borrowing
costs for households and companies
• Find short run impact on consumption and short
and long run impact on investment
• Estimates of impacts on costs are upper bound as
structure of portfolios and of relative prices may
change if regulations significantly tighten
• When capital adequacy standards tightened by 1
pcp, banks contract balance sheets by 1.2 per cent
and reduce riskiness of portfolio, with their risk
weighted assets falling by 1.6 per cent - results
contrary to the Modigliani Miller theorem of
irrelevance of the debt equity choice
• Caution needed with rapid regulatory
tightening owing to immediate impact on
margins
• Positive net benefit from regulatory
tightening, with a 2 to 6 percentage point
increase in capital and liquidity ratios
increasing welfare, depending upon
assumptions
• Separate tightening of capital or liquidity
ratios offers benefit up to 10 percentage
points
References
• Barrell R, Davis E P, Fic T, Holland D,
Kirby S, Liadze I (2009), "Optimal
regulation of bank capital and liquidity:
how to calibrate new international
standards”, FSA Occasional Paper No 38