Credit channel

Document Sample
Credit channel Powered By Docstoc
					               Credit channel (credit view)
 “Large fluctuations in aggregate economic activity sometimes
  arise from what appears to be relatively small impulses”
  (Bernanke, Gertler and Gilchrist, REStat, 1996). In the EMU, an
  increase in the official interest rates of 100 b. p. results in a
  reduction of real GDP in the range of 0.3-0.7 points for 2 years
  (The Monetary Policy of the ECB, 2004).
 Particularly strong is the effect on investment expenditure. But
  the monetary policy influences the short-term interest rates.
  Hence, it should have a relatively weaker influence on (real)
  long-term rates. So, how is it that monetary policy apparently
  has large effects on purchases of long lived assets such as
  housing and production equipment? In fact, the most rapid (and
  in percentage terms, by far the strongest) effect of monetary
  policy is on residential investment. Moreover, also the problem
  of timing is puzzling: the interest spikes are largely transitory;
  yet their effects seem to last rather long.
 Many economists (Bernanke, Cecchetti, Greenwald, Hubbard,
  Stiglitz, and others) have argued that monetary policy has direct
  effects on aggregate spending that do not operate through
  traditional interest rate (or exchange rate) channels. They have
  focused on credit markets as playing a critical role in the
  transmission of monetary policy, strengthening its efficacy. If
  credit markets can be important for the USA, where the
  dominant role belongs to the capital markets, it should be much
  more important for the euro area, where the banks are the
  dominant players. The monetary policy changes the credit
  market conditions and these amplify and propagate the effects of
  initial monetary shocks by operating along a bank lending
  channel and a broad credit channel (also known as financial

 If credit channels are important, then the level of real interest
  rates and the stock of money may not provide a sufficient
  indicator of the stance of monetary policy. The money stock is a
  much less accurate measure of the credit flows. And credit
  shocks may play an independent role in creating or in
  amplifying economic fluctuations, in addition to the effects of
  the traditional money channel. “The tightness of monetary
  policy may not be well measured by real interest rates; credit
  availability may be important” especially in the case of credit
  rationing. “The relevant interest rate (…) is the loan rate, and
  there may be marked changes in interest spreads – the difference
  between T-bill rates and loan rates” (Stiglitz and Greenwald, p.
  298). The spreads usually increase with tight money.
 The starting point of the credit channel theory is the recognition
  of the imperfect functioning of the markets for loans, bonds and
  equities. The imperfections are brought about by the
  pervasiveness of the phenomena of asymmetric information and
  of the connected agency problems that can result in large gaps
  between the lenders’ expected returns and the borrowers’ costs
  of funds. (Stiglitz and others). Of course, these consequences
  appear as a violation of the Modigliani-Miller theorem.
 The credit channel main proposition is that: “the direct effects
  of monetary policy on interest rates are amplified by
  endogenous changes in the external finance premium”
  (Bernanke and Gertler, JEP, 1995, n. 4, p. 28). The mechanism
  is: the central bank increases the interest rate for bills (and for
  bonds?); there follows a reduction of bank reserves (and
  deposits) and so there is an additional influence on the
  availability and on terms (maturities, collaterals, and so on) of
  new bank loans. This fact paves the way to the external finance

    1.The direct effect comes from the bank lending channel
                        (deposits  loans).
According to the credit view, it is necessary to distinguish
between different non-monetary assets instead of aggregating
them into a single group (bonds). In particular, it is necessary to
distinguish between bank and non-bank sources of funds and
between internal and external financing. The credit channel is an
enhancement mechanism, not a truly independent or parallel
channel: a rise in i may have a much stronger contractionary effect
if the firms’ balance sheets are already weak.

 Credit channel: 1. Bank lending channel:
   M  bank deposits  bank loans 
                I Y
it is based on the view that banks play a special role in the
financing of small and medium size firms.

  2. Other (indirect) effects come from the balance sheet
A) banks’ balance sheet: value of assets (bonds and loan
portfolio) banks’ net worth loans.
B) borrowers’ balance sheet: lower profitability or higher
probability of default: 1). Interest rates firms’ covered ratio
(ratio of interest payments to the sum of interests and profits)
expected returns loans. 2). Value of assets value of possible
collateral loans from any external (not only banks) fund
provider. So a restrictive monetary policy leads to increases in
the cost of external funds. It can also lead to credit rationing (as
we shall see later on). Detailing a bit more (Mishkin, 1995):

 2. Balance sheet channel (broad credit channel):
 M  i Pe Adv. Sel& Mor. Haz.
          loans   I  Y
                             and also

M  i cash flow  Adv. Sel.& Mor.
      Haz.loans   I  Y
the firms’ balance sheets are deteriorated by a monetary
contraction both because it causes a reduction of the net worth of
the firms (and so of borrowers’ collateral and equity stake) and
because it reduces their cash flows.
 3. Liquidity effects on consumer expenditures (durables and
  housing) channel:
     M  iPe  financial assets
       likelihood of financial distress 
       consumer durable and housing
              expenditure  Y
balance sheet effects can operate through their impact on
consumption expenditure. If consumers expect a “higher
likelihood of finding themselves in financial distress, they would
rather be holding fewer illiquid assets like consumer durables or
housing and more liquid financial assets” because, in the case they
should need to sell assets in order to raise cash, “stocks or bonds
can more easily be realized at full market value” (Mishkin p. 9).

Possible reply to a question raised before: Why short-term
interest rate increases appear to have large effects on investment
and real GDP? A possible explanation, based on asymmetries of
information, credit channels and risk-averse behaviour of firms
and banks, could follow these lines (Stiglitz and Greenwald, 2003,
pp. 132-35):

   The supply of loans by a bank is a function, among other
                                                    LS    
    variables, of the value of the bank’s capital:       0
                                                    K     
    and of the borrowers’ (firms and households) capital:
     LS           
                0  . The last inequality derives from the fact
     K F          
    that the higher K F the lower the probability of default, and
    hence the more attractive is lending.
   The maturity transformation function of the banks usually
    results in borrowing short and lending long. Especially in the
    (admittedly extreme) case of large mismatching, an increase
    in the nominal interest rate has a big adverse on the (real, not
    only nominal) value of the bank’s capital. This fact shifts to
    the left the loan supply curve.
   Moreover, the curve can shift even more because the nominal
    rate increase could have the effect of reducing the value of
    the assets held by the firms (and households) and hence their
    capital value KF.
   Other depressive effects could come from the side of the
    demand for loans. In fact, the reduction of KF may induce
    the firms to be more risk averse and so less willing to
    undertake investment (they may even decide to reduce their

  production). If so, the demand for funds will decrease. Of
  course, different firms (households) are affected differently.
  Hence, changes in nominal rates (even when there is no
  change in the real rates) may bring about very relevant
  distributive consequences: some gain at the expense of
  others (See also: Cecchetti, Fed of St. Louis Review, 1995).
 It is possible that the redistribution has the effect of lowering
  the real net worth of firms and of increasing that of
  households. “Should that not be a wash? The answer is
  clearly no, and at least part of the reason has to do with the
  equity constraints (…) If firms and banks have a lower net
  worth, they are less willing to engage in risky investment and
  lending. If there were perfect information, the households
  could simply undo the changes that the markets has wrought;
  that is firms and banks could issue more equity. However, the
  very nature of the turmoil associated with large changes in
  market values is that such changes create large asymmetries
  of information. (…) Investors will know that those banks
  most willing to restructure, to issue new equity, are those
  which are either in absolutely the worst shape (…) or those
  overvalued by the market. In either case, rational investors
  will be reluctant to provide new equity” (Stiglitz and
  Greenwald, cit. pp. 135-36).
 Note: In the interest rate channel (IS-LM and others), the
  monetary policy actions are important (only) as they affect
  aggregate investment forcing a movement along an
  unchanged marginal efficiency of capital schedule. When i
  increases, the least profitable projects are no longer
  undertaken (there are no externalities or capital market
  imperfections). The most profitable projects continue instead
  to be funded and hence there are no direct efficiency losses.
  On the contrary, the credit view takes account also of the
  distributional consequences of the monetary actions. It may
  happen that credit may be denied to a firm just because its
    current net worth is low, irrespective of the social return of
    its investment project. Hence, the shape of the aggregate
    marginal efficiency of capital schedule is a function of the
    debt-equity ratio  violation of the MM theorem (Cecchetti,
   Summary. Higher interest rates affect not only the demand
    for loans but also the supply of loans: 1. by decreasing the
    value of the long term bonds in the banks’ portfolios; 2. by
    increasing the probability of default (the firms have to pay
    higher interests and have a reduction in their capital values);
    3. by increasing the uncertainty that the banks have to face
    both at the micro level (they are imperfectly informed on
    each existing borrower’s asset position) and at the macro
    level (they may fear that the increased rates will result in an
    economic slowdown).

     Bank lending channel: Explanations and comments

 The bank lending channel follows from the observation that
 bank credit may be “special”, that is, have no close substitutes
 because of information advantages banks have in providing both
 transaction services and credit to businesses. Small firms, in
 particular, may have difficulty in obtaining funds from non-bank
 sources and also from other banks if the previous ones have
 reduced their loans. Hence, small firms can greatly suffer from a
 credit crunch when the banks face increases in the non-
 performing loans and perceive greater risks of firms’ failure. In
 such a situation the banks are in fact likely to react by drastically
 reducing the supply of loans, especially to small firms.

 In fact, credit is based on information. Its price, i.e. the interest
  rate, is only a promise to pay something in the future. Promises
  are often broken. The terms on which the banks (which are risk
  averse) supply credit depend on the information about the
  probability that the loan will be repaid. The relevant information
  is highly specific (it refers to a single firm) and can be obtained
  by utilising resources that have essentially the nature of sunk
 This fact limits the number of lenders to a small or medium size
  firm. “Credit is not primarily allocated via an auction market”
  (where the central role is played by the interest rate). Instead,
  “potential lenders [have to make] judgements about the risk
  associated with various borrowers” (…) Because of the sunk
  costs recalled above, the “loans markets are inherently
  imperfectly competitive. Only a few lenders (banks) will have
  information relevant to judging the riskiness of any particular
  borrower (Stiglitz and Greenwald, p. 295).
 The bank lending channel emphasizes the special nature of the
  bank credit. While the traditional models take account of the
  effects of interest rate rises on money demand, on consumption
  and on investment, the bank lending channel emphasizes that
  there are effects also in the supply of credit (the asset side of the
  banks’ balance sheets).
 The increase in interest rates and the reduction in banking-sector
  reserves, brought about by a monetary restriction, have a great
  influence on the supply of credit because “the direct effects of
  monetary policy on interest rates are amplified by endogenous
  changes in the external finance premium” that lenders charge to
  borrowers. The premium is defined as the difference between
  the cost of funds raised externally (loans, bonds, equities) and
  that of funds generated internally (by retention of earnings).

 The premium “reflects [the existence of] imperfections in the
  credit market”; it is due to the lenders’ expected costs of
  evaluation, monitoring and collection. It is a “lemon” premium
  required by lenders who know that, when providing funds, they
  do not have the same information as the borrower and also that
  they are taking the moral hazard risk of borrower’s decisions
  that can damage them. “A change in monetary policy that raises
  or lowers open-markets interest rates tends to change the
  external finance premium in the same direction” (B.S. Bernanke
  and M. Gertler, Inside the black box: the credit channel of
  monetary policy transmission, J. of Ec. Persp.,fall 1995, p.
 The decrease in reserves will lead banks to reduce lending if
  they cannot easily switch to alternative (f.i. CDs) forms of
  finance or liquidate assets other than loans. Perhaps, they have
  the possibility of switching to alternative forms but, since they
  do not face a perfectly elastic demand for their liabilities, they
  would face an increase in the cost of funds. This increase
  “should shift the supply of loans inward, squeezing out bank
  dependent borrowers and raising the external finance premium”
  (ibidem, p. 41).

 Key to the bank lending channel is: 1) the lack of close
  substitutes for deposit liabilities (this seems to be important for
  the American banks of small dimension which seem to be less
  liquid -as gauged by the ratio of securities to assets - and seem
  to be unable to raise uninsured CD finance). In periods of
  monetary tightening, CD interest rates usually increase by
  significantly more than the T bill rate (so also the banks’ balance
  sheets are worsened, and this fact increases the spread). 2) the
  lack of close substitutes for bank credit on the part of
  borrowers. It follows that a monetary policy decision changes
  both the interest rates and the quantity of loans.

 The importance of the lending channel is greater in Italy (and in
  Greece and Portugal) than in other European countries. It turns
  out that: “At the aggregate level, after a monetary restriction,
  deposits fall and banks reduce their lending”. But, contrary to
  what happens f. e. in the US, “small banks are not more
  sensitive to monetary policy shocks than large banks” possibly
  because of “closer customer relationships, owing to which small
  banks, which tend to be more liquid, smooth the effect of a
  monetary tightening on their supply of credit” (L. Gambacorta,
  Bank-specific characteristics and monetary policy transmission:
  the case of Italy, Bankit, Dec. 2001, p. 34).

 In any case, the willingness of the banks to bear risk
  significantly falls when their net worth is reduced by the
  worsened economic conditions. Consequently, being strongly
  risk-averse, the banks prefer to shift their portfolio towards
  Treasury bills and against loans.

 According to S. Van den Heuvel (Does bank capital matter for
  monetary transmission ?, Federal Reserve Bank of New York,
  Economic Policy Review, may 2002) a so called Bank capital
  channel might reinforce the bank lending channel. The thesis is:
  banks face regulatory capital requirements; the issue of new
  equities by a bank usually requires a “lemon premium” which is
  the greater the less capitalized is the bank, because it has less
  equity to absorb possible future losses. Hence, there is a failure
  of the Modigliani-Miller logic: “the banks’ lending will depend
  on its financial structure (...) When equity is sufficiently low,
  because of loan losses or some other adverse shock, the bank
  will reduce lending because of the capital requirements and the
  cost of issuing new equity” (p. 261). In fact, there is evidence
  that, after a monetary contraction, poorly capitalized banks
  reduce their loans by more than the banks with higher net worth.

       Broad credit channel: explanation and comments

 External finance is more expensive than internal finance
  (contrary to the Modigliani-Miller proposition and in accordance
  with the pecking-order theory of corporate finance) owing to the
  agency cost of lending, brought about by the information
  asymmetries and the consequent difficulties and expected costs
  faced by the lenders in monitoring the borrowers’ projects: costs
  of selection, of evaluation, of auditing and also the expected
  deadweight costs of bankruptcy and liquidation (it is costly for
  the lenders to seize either the full value of their credit or of the
  firm, in the case of default). These market imperfections and
  agency costs may characterize all credit markets. Also the banks
  are borrowers and have to pay a premium.
 The premium on external finance varies inversely with the
  borrower’s net worth and cash flows (Bernanke and others, The
  financial accelerator and the flight to quality, Restat, 1996).
  One reason is that, with lower net worth, the firm has both a
  higher probability of default and an increased incentive to
  misrepresent the riskiness of its projects. See the slide.
 A fall in the net worth, by raising the premium on external
  finance and increasing the amount of external finance required,
  reduces the borrower’s spending and production. Of course,
  some borrowers may be more vulnerable than others to changes
  in the credit conditions. Investment may be sensitive to firms’
  net worth and cash flows, and also to banks’ liquidity. This is
  particularly the case for the financially constrained firms. The
  same is true for inventories: there is evidence that, at a
  beginning of a recession, “inventories and short-term debt of
  small firms fall by considerably more than do those of large
  firms” (p.11).

 In fact, a restriction in monetary policy affects not only market
  interest rates, but also the borrowers’ financial position:
  directly because the firms may have floating rate debts (the
  increase in rates reduces their net cash flows) or because they
  suffer a reduction of the market values of the assets they could
  offer as collateral. Indirectly because their sales to the
  customers could be reduced by the monetary tightening. They
  may suffer a cash squeeze because the corporate income tends to
  fall more quickly than costs. Firms with poor access to credit
  markets (f. e. small firms) suffer a lot from the corporate cash
  squeeze. There is evidence that bank loans made above the
  prime rate drop by very much in recession, and that those made
  to small manufacturing firms fall relatively to those made to
  large firms. The constrained firms can be unable to increase
  their short-term borrowing and be obliged to cut production and
 Some researches for the EMU countries have shown that
  investment expenditures increase (decrease) with the increases
  (decreases) in sales and in cash flows (J.B. Chatelain, A.
  Generale and others, Firm investment and monetary
  transmission in the euro area, Bankit, Dec. 2001). Also banks’
  liquidity has an important influence on the quantity of loans
  offered by banks. Moreover, in Italy, smaller firms react to cash
  flow movements more than in other euro countries. A broad
  credit channel seems to be rather important: in fact, investment
  in each of the four largest countries of the euro area is found to
  be “quite sensitive to sales and cash flow movements” (op. cit.
  p. 34); its importance is much greater in Italy where the
  manufacturing firms’ reliance on bank credit is greater than in
  Spain, France and Germany: in 1997, the ratios of bank credit to
  firms’ total liability were respectively: 21.2%, 11.0%, 7.2%,
  6.2%. Hence, it seems that a broad credit channel is more
  pervasive in Italy than in the other countries.

 In the presence of strong credit channels, a rise in interest rates
  may have a much stronger contractionary impact on the
  economy if the firms’ balance sheets are already weak and the
  banks’ liquidity is rather low. Also the state of household
  balance sheet may reinforce the contractionary impact: “some
  major household purchases, notably housing, are linked to the
  conditions of household balance sheet” (Bernanke and others,
  1996, p. 6).
 A weakening of the borrowers’ balance sheets implies that the
  external finance premium may be continuing when there is a
  recession and interest rates are reduced. On the contrary, the
  stronger the borrowers’ financial position the weaker are the
  potential conflict of interests between lenders and borrowers.
  The latter can in fact finance their projects with greater amounts
  of own resources and, in any case, they can offer more
 Balance sheets matter. Variations in the borrowers’ net worth
  affect their ability to obtain credit and the cost of it. A recession,
  that lowers cash flows, or a decline in asset prices, that lowers
  net worth, will reduce credit availability and increase the wedge
  between the costs of external and internal finance. There is a
  problem of adverse selection: firms’ owners have less collateral
  to give as guarantee for the loans, so that the possible losses
  from adverse selection are higher for the lenders. And there is
  also a problem of moral hazard: firms’ owners have less equity
  stake in their firms, so that they have a greater incentive to
  engage in risky projects.
 The resulting impact on aggregate demand can generate a
  financial accelerator effect, particularly important for small
  firms. In fact, there may arise a flight to quality: in downturns
  “banks deny loans to weaker borrowers in favour of stronger
  borrowers” (Bernanke and others, p.9).

 Also the balance sheets of the banks can be worsened by
  restrictive monetary policy: an increase in interest rates lowers
  the value of the securities in their portfolios so impairing the net
  worth of the banks if, as it often happens, they have positive and
  uncovered duration gaps. Also, the banks may suffer a
  deterioration in the value of their assets owing to the greater
  risks of default on some of their loans. As a result, the ability of
  banks to attract funds may be reduced and hence also their
  capacity to make loans and their willingness to bear risks. A
  credit crunch may be the actual result.
 The importance of credit channels comes from the pervasiveness
  of the principal-agent problems in the credit markets. In these
  markets, the lender (principal) delegates control over resources
  to a borrower (agent). Then there are problems of adverse
  selection, of moral hazard and of monitoring costs. As a
  consequence, the Modigliani-Miller theorem is ruled out and
  there emerges a strong influence of the asymmetric information
  on the availability of internal resources, on the cost of the
  external finance and on the nature of the financial contracts
  actually available to the firms. These problems, in addition to
  the credit channels effects so far considered, may also lead to
  equilibrium rationing.