Finance and Banking NAKEcourse Robert Lensink and Elmer Sterken RuG

1
Finance and BankingNAKE-courseRobert Lensink
and Elmer SterkenRuG

• Lecture 1
• The Credit View

2
Course outline

• Two sections Micro (RL) and Macro (ES)
• Micro-section (1) equilibrium and rationing in
  credit markets, (2) adverse selection and
  signaling, (3) collateral as a sorting device,
  (4) financial contracting control, (5) the
  optimal number of creditors and duration
• Macro-section (a) the credit view a special
  role for loans, (b) credit frictions in general
  equilibrium, (c) the financial accelerator, (d)
  banking and monetary transmission, (e) finance
  and growth

3
Finance and Banking in Macro

• Arrow-Debreu world no relevance of financial
  structures (think of Modigliani-Miller), no role
  for banks, no money
• Macro quest for relevant imperfections
  information problems (missing markets, stickiness
  of prices) or imperfect competition and improper
  regulation
• This lecture information problems (credit
  view)
• Both imperfect information and imperfect
  competition issues lead to the relevance of
  finance, financial structures, and theory of
  financial intermediation

4
Macro General Outline

• The Credit View loans are special!
• General equilibrium models of net worth
• The Financial Accelerator
• Is private bank behavior itself relevant for
  monetary transmission?
• Empirics of the credit view
• Banking, finance, and growth

5
Basic Macro Hicksian IS-LM

• Economy can be described by the goods market
  (savings and investment), money market (money
  supply by the central bank and private banks and
  money demand), and the bond market
• Equilibrium is obtained via adjustment of output
  and the rate of return on bonds
• Money is special it can be controled by monetary
  authorities. There are no near-monies of any
  importance, so adjustment of the money stock is a
  strong instrument
• All long-term assets are perfect substitutes
  bonds, equity, private placements, loans. We can
  simply use one interest rate
• We will depart from this view and show that
  financial assets are special in particular loans

6
Critique on IS-LM McCallum and Nelson (JMCB,
1999)

• Fixed price level ? today there are micro
  foundations
• No distinction between the nominal and expected
  real interest rate ? we need an explicit role for
  inflation
• No recognition of enough distinct assets ? one of
  the subjects of this lecture
• Only short-run analysis permitted ? we will see a
  DSGE model based on the financial accelerator
• The capital stock is fixed ? we will see models
  that focus on capital decisions
• Not derivable from explicit maximizing analysis
  by rational economic agents ? we follow a New
  Keynesian strategy

7
Tobins General Equilibrium Approach to Monetary
Theory

• Why would agents substitute out of money into
  bonds only? Why not in real capital (Tobin)? Or
  in consumer durables (Friedman)?
• Agents do, so money stimulates capital, see
  hereafter for a short outline of the so-called
  Tobin model
• Why are some assets perfect substitutes and
  others not? Start from the idea that assets are
  imperfect substitutes
• Is financial behaviour modelled consistently? Are
  adding-up constraints and balance sheet
  conditions guaranteed?

8
Is money neutral? The
Tobin effect

• The idea is simple agents hold portfolios of
  real money and capital. If the growth rate of
  money supply increases, inflation rises, reducing
  the return on money and shifting the portfolio
  from money to capital
• It is simply money in the Solow growth model, but
  with strong implications
• There is no optimizing behavior of agents
  savings are exogenous. The Tobin model yields
  neutrality of money, but not superneutrality.
  Note money in the Ramsey growth model yields
  superneutrality, see Sidrauski, 1967

9
Tobin effect (2)

• Household wealth in money m or capital k (all
  lower case symbols represent per capita
  variables)
• There is a savings and a portfolio decision
• Production per capita y f(k)
• a ? k m is total wealth
• In the steady state sf(k) (1 - s)?1nk,
  derivation skipped, where ? m/k is the
  portfolio composition, s the savings ratio, n the
  growth rate of population and n n/(1 n)

10
Tobin effect (3)

• The share of real money in the portfolio is
  assumed to be negatively related with inflation
  inflation is the net negative return on money.
  Higher inflation therefore leads to a higher
  steady state capital stock
• Money M is neutral, since M does not affect the
  steady state, but not superneutral, since
  inflation does have an impact on the real
  equilibrium

11
Effects of money growth
f(k)
? 0.4
? 0.2
sf(k)
0
k
12
Tobin effect

• ? m/k depends negatively on the yield on k
  (which is fk r) and positively on the yield on
  m (which is -?/(1 ?)), so negatively on ?. A
  higher inflation therefore implies a lower ?
• This will lead to a higher k in the steady state,
  so money stimulates capital (assuming savings are
  exogenous)
• One can see this as a basic model of imperfect
  substitution between money, bonds, and equity
• Main disadvantage no micro foundation of savings
  behavior s

13
Tobin, JMCB (1969)

• General equilibrium approach to monetary theory
  real decisions should be integrated with
  financial decisions. Financial structure is
  relevant
• Start from an accounting framework that embeds
  balance sheets per sector and market clearing
  conditions
• Assume behaviour by various sectors, constrained
  by adding-up constraints and clearing conditions
• The Tobin framework provides a first idea of the
  credit view through imperfect substitution,
  although Tobin still gave a special role to money
  (and its impact on capital)
• The crucial assumption is imperfect
  substitutability of assets. But what are the
  foundations of these low degrees of
  substitutability? Problems in deriving expected
  returns?

14
Accounting framework
Sectors
Net total holdings
1
2
n
Assets
1
2
Consistency and adding-up restrictions on
portfolio models For each interest rate the
portfolio equations add up to 0 For wealth to
1 Across assets imperfect substitution
n
National wealth
Net worth
15
Money and output

• The Money view is in the core of IS-LM changes
  in money supply lead to changes in bond prices
  and so affect investment finance. Money is
  special, while loans are indistinguishable from
  other long-term assets
• Imperfect substitution with other assets leads to
  additional channels e.g. via equity or loans
• The credit view builds on the special properties
  of loans most importantly the private nature

16
A further appraisal of the money view Inside The
Black Box

• It is hard to find a quantitatively important
  effect of the neo-classical cost of capital
  variable e.g. on investment. This troubles the
  IS-LM model
• Monetary policy has an impact on short-term
  interest rates, but the main transmission (e.g.
  housing) is via the long-term rate. This seems to
  be in contradiction with basic IS-LM thoughts

17
Inside The Black Box The Four Empirical Facts

• Unanticipated monetary tightening has a
  transitory impact on interest rates, but
  sustained influence on real output and the price
  level
• Final demand falls quickly, but production
  follows with a lag, so inventories increase
• After a contraction residential investment falls
  sharply, followed by consumption
• Fixed business investment decreases, but it does
  so with a lag (and after housing and consumer
  durables)
• So how to deal with these stylized facts?

18
Inside The Black Box

• Information problems lead to a wedge between cost
  of funds raised externally and the opportunity
  costs of internally generated funds. Wedge
  External Finance Premium (EFP)
• Two channels (1) the balance sheet channel the
  role of net worth in setting the EFP, (2) the
  bank lending channel banks cannot replace lost
  deposits with other sources of funds
• In all cases there is special interest in private
  contracts. Probably the private character of
  these financial products also explains the time
  lags found in the empirics

19
The Credit View in general

• Equilibrium rationing in loan markets is likely
  this makes assets of banks even interesting as
  liabilities (money)
• Adverse selection, moral hazard, and costly state
  verification are the micro-foundations of the
  rationing problems
• Money versus credit the Bernanke-Blinder model
  is a simple illustration

20
Equilibrium credit rationing

• Some borrowers demand for credit is turned down,
  even if this borrower is willing to pay all the
  price and non-price elements of the loan contract
  (Baltensperger, 1978)
• The price ( interest rate) mechanism works at
  all times
• Contrast to disequilibrium rationing, like in the
  Availability Doctrine (Radcliffe report)

21
Two types of rationing

• Type I An agent cannot borrow the amount she
  wants at an existing interest rate. An entire
  group of applicants is excluded redlining
• Type II Some borrowers from an observationally
  identical group are able to get a loan, while
  others are not

22
Backward bending supply curve

• Expected return on a bank loan is not a monotonic
  function of the nominal rate of the loan
• Monopolistic bank offers the maximum interest
  rate R
• Monopolistic bank may decide to ration credit
  above a certain level of the interest rate the
  pool of applicants might get too risky

23
Backward bending supply
L
L2
Supply curve
L1
R
0
R
24
Information asymmetry in the credit market

• Ex ante adverse selection the contract design
  affects the group of loan applicants screening
  might help
• During the contract moral hazard the behavior
  of the applicants changes after signing the
  contract monitoring helps
• Ex post costly state verification. Auditing is
  an instrument

25
Simple example of credit in macro
Bernanke-Blinder (1988)

• IS-LM assumes money is special. Money has no
  close substitutes and can be isolated in policy.
  The bond market is redundant by Walras Law.
• Bonds include also loans and all kinds of private
  contracts and must so be considered to be a very
  rough capital market asset
• The Stiglitz-Weiss literature demonstrates that
  credit deserves more attention. So there is a
  further analysis of the bond market what is the
  role of private versus public contracts?

26
Bernanke-Blinder (2)

• Credit view credit is special loans! Bank
  supply loans and loan demand may suffer from
  net-worth problems (the balance sheet effect)
• Information problems in financial markets make
  credit special
• Credit-GNP relationship is more stable than
  Money-GNP-relationship
• There is no direct assumption on credit rationing
  (see for this type of literature the credit
  availability doctrine (Radcliffe report in the
  UK))

27
Bernanke-Blinder (3)

• There is a choice between bonds and loans, r is
  the interest rate on loans, i on bonds
• Loan demand Ld L(r, i, y), with y being GNP
• Bank balance sheet Reserves (R) loan supply
  (Ls) required reserves (t D) excess reserves
  (E) deposits (D)

28
Tobinesque overview
29
Bernanke-Blinder bank behavior

• Loan supply Ls l (r, i) D(1 - t)
• Bond demand Bb b (r, i) D(1 - t)
• Excess reserves demand E e(i) D(1 - t)
• with adding-up restrictions on l, b, and e
• Loan equilibrium L(r, i, y) l(r, i) D(1 - t)
• Deposit supply is 1 / (e(i) (1 - t) t) R
  m(i) R
• R tD

30
Money and goods market equilibrium

• Deposit demand D D(i, y)
• Equilibrium D(i, y) m(i) R, where m(i) is the
  money multiplier
• Bond market is the Walrasian market
• Goods market y Y(i, r)

31
Solution of the model

• Substitute the equilibrium solution of
  D(i, y) m(i) R into the loan market
  equilibrium L(r, i, y) l(r, i) D(1 - t) and
  solve for r r f(i, y, R)
• fi gt 0 as long as the money multiplier is not
  too interestelastic, fy gt 0, fR lt 0
• Substitute r f (.,.,.) into the IS-curve to get
  y Y (i, f (i, y, R) )
• Yi lt 0 the Commodities and Credit (CC)-curve

32
CC-LM model monetary expansion
i
LM
CC(?R)
CC
0
y
33
CC-LM-model

• CC-curve is affected by changes in reserves R,
  while the IS-curve is not
• CC-curve becomes the IS-curve again, if loans
  and bonds are perfect substitutes, or if
  commodity demand is insensitive to the loan rate
• If money and bonds are perfect substitutes the
  LM-curve is horizontal (liquidity trap), but
  monetary policy still effective in CC!

34
Comparative statics

• Expenditure shock shifts CC like IS
• Money demand shock shifts LM
• Rise in bank reserves is expansionary through LM
  and CC! But this creates an identification
  problem (see hereafter)
• Increase in credit supply shifts the CC-curve
  outward

35
The effectiveness of the credit channel implicit
assumptions

• Prices do not adjust instantaneously to offset
  changes in the nominal quantity of money. This is
  the problem of inflation persistence see the
  notes on Taylor, Calvo and Fuhrer-Moore hereafter
  
• The Central Bank can directly influence the
  volume of credit through R
• Loans and bonds are imperfect substitutes for
  both borrowers and banks

36
Inflation persistence Taylor (1)

• xt log contract wage set at t
• wt (xt xt-1)/2 is average wage faced by the
  firm it is assumed that wages are set for two
  periods. With a mark-up pt wt ?
• Assume a constant mark-up ( 0) pt wt
• Average expected real wage over the life of the
  contract is xt - (pt Etpt1)/2
• Real wage is assumed to be increasing in the
  level of economic activity xt - (pt
  Etpt1)/2 k yt

37
Inflation persistence Taylor (2)

• Use ?t Et-1pt - pt, one can show that
• pt pt-1 Etpt-1 k(yt yt-1) ?t/2
• Use ?t pt - pt-1, and we get
• ?t Et?t1 2k(yt yt-1) ?t
• So the price level is sticky, but inflation is
  not (the coefficient for Et?t1 is equal to
  unity)

38
Calvo staggered price setting intuition (1)

• 1- ? is the proportion of firms able to change
  their price in a period
• ? is the proportion of firms unable to change
  their price
• Aggregate price level
• pt (1- ?) xit ? pt-1

Price non-setters
Price setters
39
Calvo intuition (2)

• xit optimal price setting
• xit (1- ??) pt ??Et xit1
• ? 0 implies perfect price flexibility and so
  xit pt
• ? 1 implies inflexibility xit Et xit1

Price set at t
Myopic price
Desired price at t1
40
Calvo intuition (3)

• Myopic price is the price that would prevail in a
  flexible price equilibrium pt k pt mct, wher
  mc are marginal costs
• The price is a constant mark-up k over marginal
  costs
• In loglinear form pt pt mct
• Suppose that marginal costs mct are related to
  the output gap yt, for instance via wages mct
  ?yt

41
Inflation persistence Calvo (1)

• Firms adjust prices infrequently. Opportunities
  arrive by a Poisson process each period there is
  a constant probability 1 - ? that the firm can
  adjust the price. The expected time between price
  adjustments is 1/(1 - ?)
• The firm minimizes ½? ?j ?j Et(pit - ptj)2
• pit is the actual price in period t, pt is the
  optimal target price. We use that pitj ?j pit
  these prices remain constant!

42
Inflation persistence Calvo (2)

• Optimization of the loss function with respect to
  pit. Note that ? ?j?j is an infinite geometric
  series, so 1/ ? ?j?j (1 - ??). So minimization
  gives the optimal price xt
• xt (1 - ? ?) ? ?j?j Etptj or
• xt (1 - ? ?) pt ?? Etxt1
• Suppose that pt pt ?yt ?t, where ?t is a
  disturbance, and yt is output

43
Inflation persistence Calvo (3)

• xt (1 - ??) (pt ?yt ?t) ? ? Etxt1
• pt (1 - ?)xt ?pt-1. A large number of firms 1
  - ? will adjust their price each period
• Using ?t pt - pt-1 we get
• ?t ? Et?t1 (1 - ?)(1 - ??)/? (?yt ?t)
  ? Et?t1 ?yt ?t
• So this model is quite close to Taylors model
• An increase in ? decreases ? if the average
  time between price adjustments increases, output
  has less influence on inflation

44
Inflation persistence Fuhrer-Moore
(1)

• Main point inflation might be sticky instead of
  the price level
• Wage negotiations are conducted in terms of the
  wage relative to an average of real contract
  wages
• Real value of contracts negotiated at time t as
  xt - pt ?t. Define the index of average
  real contract wages in effect at time t
  vt (?t ?t-1)/2. remember
  two-period contracts

45
Inflation persistence Fuhrer-Moore
(2)

• Setting ?t agents (1) want at least a wreal
  contrac wage equal to (vt Etvt1)/2, and (2)
  compensation for the business cycle kyt
• Setting the real contract wage at the level
  ?t (vt Etvt1)/2 kyt (?t-1
  Et?t1)/2 2kyt
• We can rewrite this equation into
• ?t (?t-1 Et?t1)/2 2k(yt yt-1) ?t
• So now we have sluggishness in inflation
  adjustment

46
Evidence of the balance sheet channel

• Firms loan demand depends on collateral
  especially for small firms
• Firms financial position affects investment (in
  cash flow or investment Euler models for
  instance)
• External finance premium increases during a
  slump business cycle dependent

47
Bank lending channel

• One of the empirical problems is identification.
  It can be supply and/or demand effects
• Monetary policy might affect the supply of
  credit, but lower output can also decrease demand
• Bank-dependent borrowers might be rationed for
  instance small firms
• Crucial banks must meet some kind of
  imperfection to attract additional funds
• We turn to this issue in Lecture 5

48
Conclusions credit view

• Information asymmetry makes credit special
  probably a low substitution of loans with bonds
• Credit view includes the bank lending channel and
  the balance sheet channel
• Credit view affects both firms, households and
  banks behavior
• The real impact depends on the degree of price
  stickiness (e.g. Calvo staggered price setting)