FINANCIAL THEORY AND BRUNO ĆORIĆ: PRACTICE THE FINANCIAL : CONCEPT AND CHALLENGES

171 35 (2) 171-196 (2011)

1 * 2 **

stions. Any remaining errors are the responsibility of the author. Any remaining errors are the responsibility of the author. stions. Pugh, Peter Reynolds, Bill Russell, Ahmad Seyf and two anonymous referees for their many helpful sugge- Pugh, Peter Reynolds, Bill Russell, Received: November 4, 2010 This paper is based on the author’s PhD dissertation at Staffordshire University. I thank Nick Adnett, Geoff Adnett, Geoff I thank Nick University. PhD dissertation at Staffordshire This paper is based on the author’s

challenges effect: concept and concept effect: [email protected] ** UDC: 336 University of Split, Faculty of , Split University of Split, Faculty Review article Accepted: April 28, 2011 Accepted: * BRUNO ĆORIĆ, PhD

JEL: E32, E44 The financial The - - nan- nancial nancial accelerator effect. nancial markets during this period. nancial markets into a general equilibrium nancial markets into a general equilibrium a declining economy, but can them- ections of uence short-run aggregate economic activity was sugge- uence short-run aggregate since the have drawn atten- nancial crisis nancial markets, where the actions of all economic agents pro- and ex- nancial markets in real economic activity was recognized int roduct ion system functions smoothly – and smoothly enough to justify abstracting from fi system functions smoothly – and smoothly Twenty years after, the severe conse- nancial considerations (Gertler, 1988:559). quences of the worst fi cial markets are not just passive refl economic activity. The potential for the selves be a major factor depressing real active role of fi literature over the last two decades. tensively investigated in the macroeconomic on the fi Most of these researches have focused the partial equilibrium analyses about in- This literature essentially incorporates formation-based imperfections on fi information-based imperfections in fi framework. The idea that the asymmetric nancial markets can infl Bernanke (1983) argued that the credit sted by Bernanke (1983). In particular, mainly from the autonomous reaction squeeze during the great depression arose of the banking sector to increased real costs of intermediation caused by worse- ning asymmetric information problems in the fi In other words, the credit squeeze was not just the passive response of decreasing demand for loans due to the decline in economic activity. Despite this early work, it was only after Bernanke and Gertler (1989) formalized this idea in a general au- equilibrium framework that it attracted the broader attention of the economic dience. Bernanke and Gertler (1989) constructed a general equilibrium model with incomplete fi In 1988 Mark Gertler summarized the stance of macroeconomic literature in the In 1988 Mark Gertler summarized the theory presumes that the fi following sentence; Most of macroeconomic to the idea that episodes of deteriorating tion of a broader economic community burdens and falling asset in fi credit conditions, growing debt 1 Abstract markets nancial fi in asymmetry of information on the role concentrates This review nd that We fi uctuations. fl of short-run output and propagation cation in the amplifi princi- rst a consistent, fi it is known, provides effect, as nancial accelerator the fi - between fi of the relationship for the analysis theoretical framework ple based, provides a plausible uctuations. It also output fl and short-run nancial markets rst principle-based crisis, and fi causes of the recent of the proximate explanation during this crisis by for the credit policy measures taken theoretical background the Despite the theoretical plausibility, scal authorities. and fi many central banks accelerator nancial fi about the economic importance of the empirical evidence to expand existing weak. We also suggest two new aspects effect is still relatively further research. accelerator effect, which call for nancial concept of the fi nancial ac- fi nancial markets imperfections, information, fi Keywords: asymmetric cycles celerator, business

FINANCIAL THEORY AND BRUNO ĆORIĆ: PRACTICE THE FINANCIAL ACCELERATOR EFFECT: CONCEPT AND CHALLENGES 35 (2) 171-196 (2011) 172 FINANCIAL THEORY AND BRUNO ĆORIĆ: PRACTICE THE FINANCIAL ACCELERATOR EFFECT: CONCEPT AND CHALLENGES

173 35 (2) 171-196 (2011) (1) (2) ed cient nancial ciently large ciently

nancial accelerator nancial ). ) and strength of the fth section discusses fth section discusses φ are amplifi uctuations cient,

nancial accelerator is used for , output variance is, 2 is the autoregression coeffi φ will result in infinitely large variance, which is variance, will result in infinitely large

EFFECT

presents the theoreti- Section two ve sections. 1 cation and propagation mechanism, which aims to cation and propagation mechanism, which aims is less than one cient symbolizes economic shocks. Under the assumption that symbolizes economic shocks. Under the

t u is necessary to keep the variance of time series generated by an AR(1) process is necessary to keep the variance of time series generated by an nancial accelerator effect. Section three discusses different Section three discusses accelerator effect. nancial ACCELERATOR

uctuations have usually been considered a result of various uctuations have usually been output fl short-run in which and rst principles, stands for growth rates of output. t y FINANCIAL ∆ and propagation mechanisms suffi cation of economic shocks

section concludes. challenges this literature faces. The last ndings and THE

The assumption that There are number of different univariate models that strive to find the best statistical representation for short- There are number of different run properties of output time series. The AR(1) model is chosen in order to make the exposition simple and The run properties of output time series. because it has become a standard reference point for these models. Furthermore, Hess and Iwata (1997) found at replicat- AR(1) process is at least as good as, if not better than, other widely used nonlinear models that an ing business-cycle features of the US GDP. 2 1

2 Short-run output fl propagation mecha- are further transmitted by different economic shocks, which of output move- ways of thinking about the generation nisms. One of the common (see for example: period is an autoregressive (AR) process ments over a short-run 2001), Blanchard and Simon, ceed from fi ceed not observed in the output data. positive and finite. This assumption should not be considered as restrictive since that kind of variance and the This assumption should not be positive and finite. is commonly observed in output data across countries (see, for exam- size of the autoregressive coefficient autoregression coefficient Namely, ple, Ćorić, 2011). where volatility depends on the size of econo- Consequently, the size of short-run output deviation, mic shocks (size of economic shocks standard the regression error term follows a normal and that the absolute the regression error term follows a normal of the autoregression coeffi coeffi propagation mechanism (size of autoregression Identifi whose value measures the persistence of the effects of economic disturbances on whose value measures the persistence of output. The error term This study reviews and discusses a growing literature on the fi a growing literature reviews and discusses This study and propagated due to information-based credit market frictions. market credit due to information-based and propagated in fi review is organized effect. The for the fi cal rationale fi transmission mechanism. The accelerator in the monetary the fi ways to model this effect. The fourth section deals with the role of the fi with the role of fourth section deals this effect. The ways to model to explain short-run volatility observed in macroeconomic time series has been the to explain short-run volatility observed The term fi main issue of literature. the economic shocks amplifi

that arises that arises rst one is rela- Where economic Where economic 3 nance premium is nance premium is nance premium nancial outcome and might or explicit collateral require- 4 nancial accelerator. This relation nancial external fi In the market imperfections. nancial rms take too much debt in relation to and the net worth is com- nance premium and agents’ nancial outcome and the greater their incentive cient stake in the fi nancing, borrowers will be more eager to undertake rm (Bernanke, Gertler and Gilchrist, 1999:1345). The Gertler and Gilchrist, 1999:1345). The rm (Bernanke, ned as: the sum of liquid assets plus collateral value of assets plus collateral the sum of liquid ned as: rms’ losses in the cases when the project’s return is low are economic agents’ net worth net agents’ economic funds raised externally and opportu- as: the difference between the cost of ned An implicit collateral requirement presumes borrower’s assets which are not explicitly assigned as collat- An implicit collateral requirement presumes borrower’s Although the formulation with external finance premium is most frequently used it is not the only possible Although eral to a lender but which the borrower will lose in the case of default. For example, the lender’s restriction eral to a lender but which the borrower will lose in the case of default. For example, the lender’s that the maximal amount of a loan provided to a firm cannot exceed some percentage of its value, or some percentage of its asset value. 3 approach, see below. 4 vergence of the between borrowers and lender should decrease (for details vergence of the interest between borrowers explain how relatively small economic shocks can have large and persistent ef- and persistent large can have shocks economic small how relatively explain fi due to activity economic on aggregate fects the interplay it relies on (1989) formulation, and Gertler original Bernanke between due to asymmetric information between lenders and borrowers. between lenders and information due to asymmetric ments reduce this divergence of interest between lenders and borrowers; collate- ments reduce this divergence of interest the borrower participates in the project, ral, or the level of own funds with which in the case of low project returns. This increases the costs the borrower will face to undertake riskier projects that are seen way they will make borrowers less eager of view. So, as the borrowers’ net worth as unfavourable from the lenders’ point of low returns increase and, hence, the di- grows, the costs they have in the case on the theory of credit and the role of collateral in loan agreements see on the theory of credit rationing and the Bester, 1985; and Besanko and Thakor, for example: Stiglitz and Weiss, 1981; if fi 1987). According to another explanation, equity they will not have a suffi the agents’ net worth in relation to the therefore not behave diligently. The higher debt, the higher their stake in the fi to behave diligently, hence, lower the divergence of the interest between bor- rowers and lenders (Holmstrom and Tirole, 1997). defi the fi nity costs internal to the external fi relationship between as the keystone of the fi monly acknowledged illiquid assets less outstanding obligations; and the external fi obligations; and less outstanding illiquid assets agents’ net worth is defi agents’ net (Bernanke and Gertler, 1990). rests on two presumptions to the project, of his own wealth the borrower contributes First, the less the amount supplier of the external will diverge from the of the the more his interests The fi basic explanations for this presumption. funds. There are two in loan agreements. credit rationing and the role of collateral ted to the theory of In brief, in the case of debt fi riskier projects. That is, projects that have a high probability of large return, but riskier projects. That is, projects that have borrower’s perspective these projects are also those offering low returns. From the preferred since the fi limited to zero by legal regulation (limited liability assumption). From the len- limited to zero by legal regulation (limited since they bear all, or most of, ders’ point of view, these projects are unfavourable Implicit the costs in the case of low project returns.

FINANCIAL THEORY AND BRUNO ĆORIĆ: PRACTICE THE FINANCIAL ACCELERATOR EFFECT: CONCEPT AND CHALLENGES 35 (2) 171-196 (2011) 174 FINANCIAL THEORY AND BRUNO ĆORIĆ: PRACTICE THE FINANCIAL ACCELERATOR EFFECT: CONCEPT AND CHALLENGES

175 35 (2) 171-196 (2011) nancial ac- nance, hence also gure 1). A change nancial asset, etc.), What types of costs age- What types 5 nance premium makes bor- nance premium. This inverse nance premium. This costless, and the costs of the same trans-

ed and propagated by the fi during the expansio- cult and/or expensive during than nance premium is inversely related to economic agents’ net worth, nance premium is inversely related to economic their net worth. In so far as nance premium, are inversely related to nancial accelerator effect can be described as follows (fi nancial accelerator effect can nance premium. Consequently, due to higher costs, and/or reduced ability to nance Agency costs are a broad range of different costs that arise in all economic transactions where the parties parties costs that arise in all economic transactions where the Agency costs are a broad range of different 5 Second, in cases when the borrowers have superior information about projects’ about projects’ information superior have borrowers when the in cases Second, that actions unobserved to take or abilities etc.), (value, riskiness, characteristics of interests incompatibility a greater projects’ return distribution, can affect costs. increases agency and lenders between borrowers involved in those relations have different interests and where their actions are not observable, or at least not interests and where involved in those relations have different They are the wedge between the so-called first best and costlessly observable by all transaction participants. between costs of the transaction that takes place between two That is, the difference the second best solution. parties in situations in which the information is equally shared and action in the situation where information is not equally distributed between parties. The fi ncy costs would include depends on the way the information asymmetry in the information asymmetry on the way the would include depends ncy costs the assump- a model is built on below). In brief, if is modelled (see credit market will include the agency costs are solvable, then problems tion that information to spend to overcome that the lender would be “forced” costs of all real resources is that those pro- the other hand, if the underlying assumption these problems. On and credits are rationed or at least not completely solvable, blems are not solvable, suboptimal allocation agency costs will include the costs of in equilibrium, then of interests between So, the greater the incompatibility of funds in the economy. spent on monito- the greater the extent of real resources lenders and borrowers, rationing in the eco- and/or the greater the extent of credit ring, selecting, etc., the higher the the greater the incompatibility of interests, nomy. Consequently, the lower availability of credits. and/or Economic disturbances that can be amplifi in aggregate economic activity causes a change in economic agents’ net worth in aggregate economic activity causes them. Due to imperfect information, the because of a positive correlation between able to raise external fi terms under which economic agents are rowing more diffi in investment, spending and produc- nary phase. This in turn exaggerates swings any negative economic shock that might tion over business cycles. For example, net worth would also increase the external lead to a decrease in economic agents’ fi spending and production will borrow, the overall level of agents’ investments, even further. decrease. In turn, that will depress the economy that cause: changes in the value of econo- celerator mechanism include all shocks mic agents’ liquid assets (change in cash position, short-term fi de- such as, a change in productivity, a change in caused by a crease in supply, a decrease in foreign demand, etc.; changes in the value the external fi the external fi agents’ net worth over business cycles the procyclical behaviour of economic external fi implies countercyclical behaviour of the external fi relation between output changes and the ect: the ASYMMETRY in which nancial markets nancial accelerator effect.

rst two approaches, established nancial accelerator effect refl nancial accelerator effect into general nancial accelerator INFORMATIONAL

MODELLING nancial market imperfections.

IN

STRATEGIES

uenced by fi the problem of asymmetrically distribu- nancial markets can overcome 1 DIFFERENCES

MODELLING IGURE

In general, it is possible to distinguish three major types of modelling strategies In general, it is possible to distinguish fi used by researchers to incorporate the 3.1 3 the fi The different modelling approaches to the asymmetric information problem variety of different situations in which different fi between borrowers and lenders may emerge; toward the question whether partici- this problem can appear; different attitudes pants in fi of borrowers whose economic activities ted information or not; and different types can be infl equilibrium models (the main differences among these approaches are summari- sed at the end of this section in the table 1). The fi The arrows symbolize the mechanisms (channels) through which the financial accelerator opera- symbolize the mechanisms (channels) through The arrows 1 represents Arrow events and their consequences for the economy. tes, and the boxes represent net worth. economic activity and agents’ between changes in aggregate a positive relationship between net worth and the size of the exter- an inverse relationship 2 represents Arrow In turn, between the external financial an inverse relationship 3 represents Arrow nal finance premium. pro-cyc- represent arrows the return Finally, and investment, spending and production. premium economic activity. lical feedback into aggregate by Bernanke and Gertler (1989) and Kiyotaki and Moor (1997), consider informa- tional asymmetry on credit markets as the cause of the fi of economic agents’ illiquid assets, that can arise due to either a change in interest change in to either a arise due that can assets, illiquid agents’ of economic of eco- by actions caused can be both of which prices in the asset a change rate or economic about future in the expectations makers or by changes nomic policy can also be obligations agents’ outstanding etc.; change in economic performance, outstanding the assumption that rate, under a change in the interest caused by rates. to variable interest loans are subject F The third approach, established by Greenwald and Stiglitz (1993) considers infor- mational asymmetry on equity markets and managers’ as the cause of this effect. The fi nancial accelerator effect nancial accelerator The fi

FINANCIAL THEORY AND BRUNO ĆORIĆ: PRACTICE THE FINANCIAL ACCELERATOR EFFECT: CONCEPT AND CHALLENGES 35 (2) 171-196 (2011) 176 FINANCIAL THEORY AND BRUNO ĆORIĆ: PRACTICE THE FINANCIAL ACCELERATOR EFFECT: CONCEPT AND CHALLENGES

177 35 (2) 171-196 (2011) nancial nance that cult to build and hi- ows, balance sheet position, management, ows, balance sheet position, nancial markets can overcome the problem of overcome the problem markets can nancial nance premium will in this case take the form of markets, on credit which concentrate rst two approaches, to the borrower compared to the use of internal nance expensive nance premium. nance where these costs are not present. The difference between these two sour- nance, On the other hand, the assumption that asymmetric information problems between On the other hand, the assumption that asymmetric information problems between borrowers and lenders are not solvable implies that the borrowers would not always be able to obtain credit, or at least not the amount considered optimal from their point of view. In particular, in circumstances when lenders are not able to overco- accelerator effect would be extremely complicated, very diffi accelerator effect would to use one particu- strategy adopted by researchers has been ghly intractable. The problems as representative of all sce- lar formulation of asymmetric information or not those problems are solvable. As a narios and to assume a priori whether formulations of the agency costs and the result, different models have different external fi problems between borrowers and The assumption that asymmetric information would use various techniques to deal lenders are solvable implies that the lenders selecting among borrowers; monito- with this problem: for example, screening; ring (inspection of borrowers’ cash fl relationship with borrowers; and realized return, etc.); engaging in a long-term as a minimum solvency ratio or a mini- enforcement of restrictive covenants such instruments requires lenders’ real re- mum cash balance, etc. The use of all these lenders have to be compensated in equili- sources and entails real costs for which these techniques are able to reveal all the brium. Therefore, even in the cases when their implementation makes ex- hidden or unobservable borrower characteristics, ternal sources of fi fi of the higher interest rate borrowers will ces of funds will be expressed in the form be asked to pay. The external fi the difference between the interest rate on the external source of fi nts with low net worth would be charged compared to the interest rate those age agents would be charged in the situation where their net worth is high. In particular, the asymmetric information problem can emerge because borrowers’ information problem can emerge In particular, the asymmetric or any combination or project riskiness, or project quality, intentions, or honesty, transactions take place is not observable to lenders before of these characteristics has taken place, information problems). Once the transaction (ex-ante asymmetric return or duration, able to observe borrowers’ actions, project lenders may not be information problems) to repay loans (ex-post asymmetric or to force borrowers possible scenarios in A model that would include all the (Jaffe and Stiglitz, 1990). of the fi information problem may arise as sources which an asymmetric arise due to the variety of situations in which informational asymmetry between asymmetry informational in which situations variety of due to the arise and differen- may be evident, on the credit markets and investors ultimate savers to the que- authors with respect taken by these underlying assumption ces in the fi participants in stion whether or not. distributed information asymmetrically Differences between the fi the between Differences nance cation procedure cation procedure, cation procedure is rms, and the overall rms will not be able to rms will not be able cation procedures only cation hypothesis. This will be rms’ internal funds rms are willing to pay, but rms are are equal to the diffe- ts of fraud cation in cases in which borrowers ciently, that is, below the level where ciently, that is, below nancial accelerator effect, Bernanke and Ger- nancial accelerator effect, Bernanke and nance premium and net worth comes from the nance derive from fraud. The way lenders ts borrowers ts is to increase the proportion of borrowers in the group of rms that declare bankruptcy, borrowers have the incentive to re- rms that declare bankruptcy, borrowers borrowers’ incentive to report false ts of fraud stayed unchanged, costs of the external funds, and will contain rms’ internal funds and the marginal they will be unavailable. The marginal productivity of fi marginal productivity unavailable. The they will be (1989) approach literature has adopted Bernanke and Gertler’s The majority of the al., 1999; Aoki et al., and Fuerst, 1997; Bernanke et (see for example: Carlstrom and Dib, 2008; Portes 2006; Gertler et al., 2007; Christiansen 2004; Elekdag et al., Magud, 2010; Freedman et al., 2010; and Ozenbas, 2009; von Heideken, 2009; Calvalcanti, 2010). To model the fi costly state verifi tler (1989) applied Townsend’s (1979) caused by asymmetric information hypothesis thesis states that the problems the fact that lenders are not able to ob- between borrowers and lenders arise from cost-free. The borrowers can report serve borrowers’ project return realizations mitigate debt obligations and retain the false project results (declare default) to project returns. Due to this possibility difference between reported and realized verifi lenders are forced to implement return of return verifi declare bankruptcy. Since the implementation a random verifi costly, to minimize costs lenders implement caught to cheaters. An inverse rela- which constitutes a realistic threat of being tionship between the external fi means that the borrower has less to lose in point that the low borrower’s net worth apply return verifi the case of fraud. Namely, since lenders to some of the fi borrowers will have when they declare port false project results. The costs net worth plus reported project return. bankruptcy are equal to the value of their benefi On the other hand, the potential (expected) project results will rise. To prevent potential losses due to dishonest behaviour, project results will rise. To prevent potential losses due to dishonest behaviour, lenders should reduce the benefi can reduce these benefi borrowers who declare bankruptcy to which the returns verifi me asymmetric information problems by screening, monitoring and other techni- and other monitoring by screening, problems information me asymmetric Weiss, and (Stiglitz but by quantity by credit not ration they might ques, even though able to obtain funds will not be cases, some borrowers 1981). In these this as- value (NPV). Under a positive net present projects have their investment the fi cost even less than external funds will sumption, but the fi of external funds, the marginal costs higher than fi to equalize this difference. Hence, borrow enough funds rence between realized and reported project returns multiplied by the probability the they will not be caught. So, in a case in which borrowers’ net worth decreases, the costs they will have if they report fake project returns will also decrease. Since potential benefi economy, will be forced to produce ineffi economy, will be forced the external fi meets marginal costs. In those cases marginal productivity premium will take the form of the difference between the marginal value of the the form of the difference between the premium will take fi opportunities. resources and missed production the costs of all the wasted

FINANCIAL THEORY AND BRUNO ĆORIĆ: PRACTICE THE FINANCIAL ACCELERATOR EFFECT: CONCEPT AND CHALLENGES 35 (2) 171-196 (2011) 178 FINANCIAL THEORY AND BRUNO ĆORIĆ: PRACTICE THE FINANCIAL ACCELERATOR EFFECT: CONCEPT AND CHALLENGES

179 35 (2) 171-196 (2011) rms’ rms’ it nancial premium, become nance will net worth. Na- rms’ rm’s share price, indepen- uctuations due to informatio- rms’ managements consider exi- rms are allowed to raise funds on a fric- rms’ net worth, in these circumstances, rms’ net worth, in these nancial accelerator effect into the general equili- nance will become less available. Consequently, from formula- nancial accelerator effect proceeds rms’ managements have “better” information about rms’ rms’ existing assets and/or investment projects, the fi rms’ existing assets and/or investment fl cation of aggregate economic markets and the procyclicality of fi nancial ra- to raise external funds. The level of credit cantly their ability rms’ true value, they consider every equity issue in the same way. The rms’ ability to obtain loans will be directly dependent on the value of the rms’ ability to obtain loans will be directly uences signifi uences brium framework, was established by Greenwald and Stiglitz (1993). Although brium framework, was established by of the external fi this formulation does not rely on the existence The third modelling strategy, which is somewhat different, but can still be conside- The third modelling strategy, which is somewhat red a way of incorporating the fi still produces amplifi nal asymmetry in fi will be applied. That is, to increase the probability that cheaters will be caught. will be cheaters that the probability to increase That is, applied. will be rates interest require higher they will lenders, costs on imposes real auditing Since economic when an adverse So, in circumstances for those costs. to compensate source of fi worth, the external borrowers’ net shock decreases Another way of modelling the fi Another way of modelling more expensive. Consequently, investment, spending and production at the aggre- and production investment, spending Consequently, more expensive. will decline. gate level asymmetry. In parti- are not able to overcome information tions in which lenders information problems Moor (1997) assumed that asymmetric cular, Kiyotaki and by any means. lenders are not able to enforce debt repayment are unsolvable since secured by the willing to lend if the loan is not completely Hence, lenders are not and machinery). In durable assets (such as land, buildings, value of borrowers’ but by quantity as well, will be rationed not just by price such an economy, credits and fi offer. Any shift in fi collateral they can infl fi mely, in Greenwald and Stiglitz (1993) from accessing the equity market. Limited tionless credit market, but are precluded the asymmetric information considerations access to the equity market is based on and Myers and Majluf (1984). In brief, in discussed by Greenwald et al. (1984), circumstances in which the fi decisions to issue shares signals “bad news” to investors. Conversely, the fi decisions not to issue shares signals “good news” to investors. In particular, inve- stors consider equity issue as a signal that the fi sting share prices to be overvalued. Since investors do not have full information about the fi dently of the fact whether or not it is truly overvalued. Hence, even in the cases consequence is that equity issue typically decreases the fi the true value of the fi tioning will be countercyclical due to the procyclical value of collateralised assets. tioning will be countercyclical due to the spending and production over business This will exaggerate swings in investment, economic shocks decrease borrowers’ cycles. So, in circumstances when adverse net worth, external sources of fi the aggregate level will decline. This ap- investment, spending and production at by Kiyotaki (1998), Iacoviello (2005), proach is adopted and further developed (2010), and Martin and Ventura (2010). Monacelli (2009), Gertler and Kiyotaki rms are rms rms’ net rms’ nancial acce- nancial nancial sour- nancial market im- Greenwald and Stiglitz’s (1993) informational asymmetry on equity markets and risk aversion of fi managers problem cannot be solved willingness to borrow rms to act in a risk-averse to act in a risk-averse rms xed obligations in fi rms’ net worth also reduce the level of rms’ production get translated into de- Kiyotaki and Moor’s (1997) rms are risk-averse, changes in fi rms are risk-averse, informational asymmetry on credit markets problem cannot be solved availability of credits rms, and through this mechanism shocks get rms, and through this mechanism shocks new shares to issue be reluctant will often rms rm may contribute to investment in production. rm may contribute to investment Bernanke and Gertler’s (1989) informational asymmetry on credit markets problem can be solved cost of credits rms want to maintain the same level of investment in produc- rms want to maintain the same level rms operate in a stochastic environment where the outcome of where the outcome environment operate in a stochastic rms rms behave in such a manner as to minimize the probability of to minimize the probability in such a manner as rms behave nancial 1 ABLE nancial Cause of the fi accelerator effect Ability of fi markets participants to overcome informa- tional asymmetry Change in net worth induce change in T manner. Finally, fi manner. Finally, every production and investment decision is uncertain. In an environment which is investment decision is uncertain. In an every production and and in which fi intrinsically stochastic when a project’s NPV is positive, fi positive, NPV is a project’s when in value increase outweigh can in share prices decrease the because and invest, (1993) Greenwald and Stiglitz NPV. Further, the project’s positive arising from that fi also assume for mana- and especially is costly for stockholders since bankruptcy bankruptcy, fi These costs induce to loss of reputation. gers, due Since fi increases the probability of bankruptcy. ces, in a stochastic world, will rather decrease be reluctant to increase borrowing, but risk averse they will fi investment in production. Decreases in (2002) and Gatti et al. (2007). lerator effect has been adopted by Arnold creases in the demand facing other fi the previous models, fi transmitted further. Overall, similar to amplify changes in economic ac- perfections will, due to net worth procyclicality, approach to modelling the fi tivity over the business cycle. So far, this Different approaches to modelling the fi nancial accelerator effect fi Different approaches to modelling the their own funds with which the fi their own funds with Consequently, if fi tion, they have to increase borrowing. Insofar as debt obligations are in the form of borrowing. Insofar as debt obligations tion, they have to increase a rise in the share of fi state-independent claims, worth position can have potentially large effects on their willingness to produce. have potentially large effects on their willingness worth position can shocks that reduce fi For example, economic

FINANCIAL THEORY AND BRUNO ĆORIĆ: PRACTICE THE FINANCIAL ACCELERATOR EFFECT: CONCEPT AND CHALLENGES 35 (2) 171-196 (2011) 180 FINANCIAL THEORY AND BRUNO ĆORIĆ: PRACTICE THE FINANCIAL ACCELERATOR EFFECT: CONCEPT AND CHALLENGES

181 35 (2) 171-196 (2011)

cant uen- nance rms) is based on nancial accelerator accelerator nancial nancial market im- nance transactions are uctuations. The general uctuations. Following Bernanke and Following nancial accelerator effect 6

of fi uenced by the same kind nance premium, which leads to a rise in housing nance premium, which leads BORROWERS nance investments in housing and purchases of other

In parti- nance some of their expenditures by borrowing. OF nancial accelerator literature is that almost all studies in studies almost all is that literature accelerator nancial

rms’ borrowing in the formulation of interactions between of interactions in the formulation rms’ borrowing a signifi mechanism will likely include nancial accelerator rms, fi rms, TYPES

banks. rm borrowers such as households and nancial markets and short-run economic fl markets and short-run nancial nancial accelerator effect emerges due to an asymmetric information rms’, since households are unable to raise alternative sources of fi rms’, since households are unable nancial accelerator effect on household spending occurs because house- nancial accelerator effect on household agents. group of economic concentrate on one eld DIFFERENT

eld concentrate on fi eld concentrate households’ spending and invest- rms’ investments should also work through For a model which simultaneously combines the financial accelerator effect on firms’ and banks’ economic and banks’ on firms’ For a model which simultaneously combines the financial accelerator effect 6 (1997). Iacoviello (2005) combines Tirole activity in a partial equilibrium environment see Holmstrom and activity in a general equilibrium model. and households’ on firms’ the financial accelerator effect bles). Consequently, the same mechanism that propagates economic shocks through bles). Consequently, the same mechanism causes a rise in house prices, which leads to an increase in homeowners’ net worth. causes a rise in house prices, which leads 3.2 of the fi feature Another this fi Gertler (1989), Greenwald and Stiglitz (1993) and Kiyotaki and Moore’s (1997) Kiyotaki and Moore’s Stiglitz (1993) and Greenwald and Gertler (1989), in this equilibrium models majority of general the large seminal contributions, fi This decreases the external fi consumption demand (spending on dura- investments and also spills over into fi to assume that the effect of the net ments decisions. It does not seem unreasonable should be even more pronounced than worth changes on households’ investments on the fi imperfect fi imperfect Recently Gertler and Kiyotaki (2010) developed a general equilibrium model in Recently Gertler and Kiyotaki (2010) developed a general equilibrium model which the fi through equity or bond issues, but are exclusively oriented towards banks as the through equity or bond issues, but are exclusively oriented towards banks as sources of external funds. ced by their net worth. Since empirically a large proportion of households’ bor- ced by their net worth. Since empirically literature has focused primarily on the ef- rowings are secured by real estate, the contributions include: Aoki et al., 2002, fect of changes in house values (seminal (2004) described the fi 2004; and Iacoviello, 2005). Aoki et al. A positive shock to economic activity effect on household spending as follows. durable by raising funds in credit markets. These fi durable goods by raising funds in credit cular, households usually fi problems between the borrowers also characterized by asymmetric information households’ ability and/or terms (households) and the lenders (banks). Therefore, hence their spending, are also infl under which they are able to obtain funds, The fi holds, as well as fi orientation of the literature toward explanation of the fi orientation of the literature relatively simple to technical concerns and keep models the intention to reduce that the activity of This strategy is not a product of beliefs maintain tractability. is not infl other economic agents a complete contrary, as Bernanke et al. (1999) emphasized, perfections. On the description of the fi role for non-fi through the behaviour of one group of economic agents (usually fi of one group of economic agents (usually through the behaviour

rms’ rms and house- nance these acti- nancial accelerator it might also affect banks’ it might also 7 nancial markets. Since banks enter banks enter Since markets. nancial MECHANISM

capital). To the their net worth (bank uenced by nancial accelerator effect in the so-called great modera- nancial accelerator and international transmission of business TRANSMISSION

nancial accelerator framework; Olivero (2010) explored the relation- nancial accelerator effect has recently been used in many different areas of nancial MONETARY

rms, there is no reason to assume that banks’ ability to collect funds and/or the to collect funds and/or that banks’ ability no reason to assume rms, there is For example, negative economic shock can increase the non-performing loans rate and/or decrease the value For example, negative economic shock can increase the non-performing loans rate and/or decrease the value 7 research, for example: Cepedes et al. (2004), Gertler et al. (2007) and Magud research, for example: Cepedes et al. of different exchange rate regimes (2010), among others, analysed properties using the fi ship between the fi this effect in the analysis of the relation- cycles; Aghion et al. (2005) incorporated Assuming that entrepreneurs raise funds ship between output volatility and growth. 4 The fi problem that constrains the ability of banks to obtain funds from depositors in depositors funds from to obtain of banks the ability that constrains problem fi (“inter-bank”) in wholesale as well as retail as well as they can go bankrupt and given that market as borrowers the deposit fi infl funds will not be costs of the worth affect banks’ net the economic shocks extent that effect based on the general-equilibrium feedback between credit and labour effect based on the general-equilibrium feedback between credit and labour market; Portes (2007), Portes and Ozenbas (2009), and Ćorić and Pugh (2011) explore the role of the fi on imperfect credit markets, Wasmer and Weil (2004) developed a theory of job on imperfect credit markets, Wasmer and Weil (2004) developed a theory of creation and job destruction that suggests the existence of a fi ability to attract funds. Why does banks’ net worth matter for attracting funds? matter for attracting does banks’ net worth attract funds. Why ability to Depositors cannot as before – information asymmetry. The reason is the same money. However, risk banks are taking while investing their observe how much implies larger costs of is costly. Higher bank net worth they know that bankruptcy higher bank net lower incentive to take risk. Additionally, bankruptcy and, hence, unanticipa- liquid bank, that is, a better ability to compensate worth means a more depositors will be a lower probability of bankruptcy. In sum, ted losses and again to put their money in will request a lower deposit interest rate more willing and/or to “contribute” to bank worth. They will also be more willing bank with high net bank has higher net in bank’s issue of new shares if that capitalization by investing in banks’ net because a negative effect of a reduction worth. This is important by banks’ intention to maintain a desired worth on loans supply can also be caused below). Unlike that of fi or imposed capital adequacy ratio (see does not consist of spending on durables holds, the economic activity of the banks Yet, since banks fi and houses, or of production and investment. gate net worth can have an effect on aggre vities to a large extent, a change in their The terms and amounts of the funds the spending, investment and production. unts under which they lend, and the amo banks collect directly determine the terms Therefore, adverse changes in their they are able to offer to potential borrowers. negatively affect households’ and fi net worth can cause credit tightening and expenditure and overall economic activity. of banks’ securities. In the case of the exogenously induced increase in interest rate banks can also suffer capi- securities. In the case of the exogenously induced increase in interest rate banks can also suffer of banks’ assets and liabilities. tal erosion due to a maturity mismatch between banks’

FINANCIAL THEORY AND BRUNO ĆORIĆ: PRACTICE THE FINANCIAL ACCELERATOR EFFECT: CONCEPT AND CHALLENGES 35 (2) 171-196 (2011) 182 FINANCIAL THEORY AND BRUNO ĆORIĆ: PRACTICE THE FINANCIAL ACCELERATOR EFFECT: CONCEPT AND CHALLENGES

183 35 (2) 171-196 (2011) ows from from ows rms’ net worth. rms’ net worth. ned as exogenou- nancial accelerator accelerator nancial A change in balan- 8 ows, that is, fl uctuation in Japan using in Japan uctuation rms. Free- accelerator, nancial effect has nancial accelerator the crises. scal stimuli during nancial accelerator effect has been consi- rms’ liquid asset fl rms’ indebtedness to the extent that corporate loans nancial accelerator effect forms the theoretical background for the nancial rms’ illiquid assets simply because the market will discount future re- rms’ illiquid assets simply nance premium and, consequently, depresses the aggregate economic nance premium and, consequently, depresses the aggregate economic the so-called of mone- uence aggregate economic activity, nancial accelerator effect has recently been used to explain the bank lending uence aggregate economic activity through policy-induced changes in loan sup- nancial market imperfections as a channel through which mea- as a channel through which monetary nancial market imperfections In the monetary literature the financial accelerator effect and the term net worth are also known as the bala- In the monetary literature the financial accelerator effect 8 and balance sheet. nce sheet effect The fi channel as well. According to the bank lending channel, monetary policy can in- channel as well. According to the bank lending channel, monetary policy can fl ply. In short, the tightening of monetary policy reduces banks’ lending capacity which, in turn, causes a decline in loan supply and/or tightening of the lending been particularly intensively used. Almost in parallel with the fi Almost in parallel intensively used. been particularly that considers information-based of literature has been a development theory there fi sures infl and Blinder, 1988, contributions include: Bernanke tary policy (groundbreaking 1993; Hubbard, 1995; Lown, 1991; Gertler and Gilchrist, 1992; Bernanke and more recent papers 1995; and Bernanke and Gertler, 1995; Kashyap and Stein, Van den Heuvel, 2005; Gertler et al., 2007; Monacelli, 2009; include: Iacovielo, 2010). The fi 2009; Gertler and Kiyotaki, is based (this chan- mechanisms on which the credit channel dered one of the main channel). lending channel and the balance sheet nel consists of the bank That is, the fi balance sheet channel. According to the balance sheet channel, a change in mone- balance sheet channel. According to the (net worth) of fi tary policy impacts on the balance sheet For example, in the monetary policy literature, the fi policy literature, the in the monetary For example, tion. Hirose (2008) investigated sources of asset price fl asset price of sources investigated Hirose (2008) tion. this model. Using the general equilibrium model with fi with model equilibrium the general Using this model. the effects of fi (2010) estimated dman et al. ce sheet affects their borrowing and, consequently, the overall aggregate econo- ce sheet affects their borrowing and, consequently, monetary policy are defi mic activity. In particular, if changes in sly induced changes in the interest rate (Bernanke and Mihov, 1998), then mone- sly induced changes in the interest rate all three components of fi tary policy changes can have effect on turns from those kinds of asset by a higher interest rate. Further, an increase in the turns from those kinds of asset by a higher interest rate may increase fi hence also the investment and spen- activity. For the same reasons the borrowing, in the same manner. ding of households, should be affected For example, an increase in interest rate should induce a decrease in the market For example, an increase in interest rate value of fi the policy-induced increase in interest have variable interest rates. Finally, since rate depresses the aggregate demand, fi due to a decline in demand for pro- existing economic activity, can also decrease increase in interest rate pushes all three ducts. Thus, the monetary policy-induced unfavourable direction. That increases the components of net worth in the “same”, external fi nancial rst channel is (see for exam- 10 nancial markets nancial uctuations even when all economic effect nancial accelerator The recent literature, however, has focused on the has focused on the literature, however, The recent 9 ed and propagated due to information-based fi ed and propagated nancial accelerator effect demonstrates that irrationalities nancial accelerator effect demonstrates activity, and that fi nancial markets in real economic tability and reduce their capital. Taken together, when the DISCUSSION

This potential effect is also known as the bank capital channel. This potential effect According to this point of view, an increase in the reserve requirement level or/and the central bank’s open an increase in the reserve requirement level or/and the central bank’s According to this point of view, market frictions. The fi myopia, are not necessary to explain the such as herd behaviour, animal spirit or active role of fi fl can have an active role in short-run output agents conform to the postulates of and optimising beha- viour. market sales will contract the funds banks have at their disposal for lending and so reduce loan supply. The market sales will contract the funds banks have at their disposal for lending and so reduce loan supply. underlying assumptions are that banks are not able to easily compensate for lost deposits and that borrowers are unable to frictionlessly substitute bank loans with other sources of finance. 9 10 5 theoretical framework in which econo- The literature reviewed offers a coherent mic shocks are amplifi on average, are less frequently (re)negotiated. Hence, this mismatch should decrease on average, are less frequently (re)negotiated. banks’ profi effects should have a decrease in banks’ increases the short-term interest rate both banks’ capital will increase the agency capital as a consequence. Reduction in negatively affect banks’ ability to attract costs of collecting funds and, in turn, collection will then cause decline in loan funds. The limitations in banks’ deposit activity. supply and the reduction in overall economic conditions, and the reduction in overall economic activity. Opinions differ about differ Opinions activity. economic in overall the reduction and conditions, The reasons. and for what supply, credit changes policy kind of monetary what the central requirements and the effects of reserve view focuses on “traditional” market operations. bank’s open potential effect of monetary policy changes on banks’ net worth (capital), and re- net worth (capital), changes on banks’ of monetary policy potential effect due to the fi in loan supply sulting change ple: Kishan and Opiela, 2000; Hubbard, Kuttner and Palia, 2002; Aikman and Vi- and Palia, 2002; Aikman Hubbard, Kuttner and Opiela, 2000; ple: Kishan 2002, 2009). Na- and Mistrulli, 2004; Van den Heuvel, leghe, 2004; Gambacorta affect banks’ capital changes in the short-term interest rate mely, to the extent that ability to attract to information asymmetry, affect banks’ they might also, due that an increase in the loans supply. This literature suggests funds and, hence, their The fi banks’ capital through two channels. interest rate might reduce in banks’ holdi- and Gertler (1995), who note an increase recognised by Bernanke increase the sensi- and derivative instruments. This might ngs of volatile securities a rise in interest to changes in the interest rate. For example, tivity of bank lending capital and negatively of securities. That reduces banks’ rate lowers the value proposed by Van den to attract funds. The second channel is affects their ability maturity mismatch between banks’ assets Heuvel (2002, 2009). It arises due to a have a shorter time duration than bank and liabilities. Namely, deposits usually bank loans are slower to adjust to changes loans. This means that interest rates on on their liabilities, simply because loans, in the interest rates than the interest rates

FINANCIAL THEORY AND BRUNO ĆORIĆ: PRACTICE THE FINANCIAL ACCELERATOR EFFECT: CONCEPT AND CHALLENGES 35 (2) 171-196 (2011) 184 FINANCIAL THEORY AND BRUNO ĆORIĆ: PRACTICE THE FINANCIAL ACCELERATOR EFFECT: CONCEPT AND CHALLENGES

185 35 (2) 171-196 (2011) - - nancial ows and ed existing ed scal authori- scal nancial interme- nancial nancial accelerator nancial nancial borrowers’ nancial nancial borrowers; ex- based rst principle and ac- rms’ net worth nancial institutions (Gertler ed the deterioration of the net amplifi nancial premium and nancial markets (Bank of England, 2007). of England, 2007). markets (Bank nancial 11 standards. Decline in house ow of credits and lending nancial borrowers face tightened and loan supply declined. nancial borrowers face tightened and institutions’ net worth were taken by monetary and fi nancial provide a theoretical background for the credit nancial accelerator can also fi a plausible also offers literature accelerator nancial scal authorities, which are often seen as old-fashioned and contradictory of scal authorities, which are often seen as It is important to stress that this is just one of the possible explanations of proximate causes. The ultimate The ultimate It is important to stress that this is just one of the possible explanations of proximate causes. nancial institutions; by lending directly to high grade non-fi nancial institutions; by lending directly to large fi and by equity injection and debt guarantees measures to facilitate credit fl and Kiyotaki, 2010). The same or similar boost fi 11 policy measures taken during this crisis by the Fed and many other central banks policy measures taken during this crisis and fi the Fed reacted to the current crisis using modern economic theory. In particular, making imperfectly secured loans to fi three types of credit policy measures: by The fi and real estate prices caused erosion of households’ and fi and real estate prices in their debt capacity. Weakening of non-fi cordingly a decline output. Decline of the aggregate econo- decline of investment, consumption and and reduction in housing prices mic activity, the ensuing rise in loans. This reduced intermediaries’ profi increased the amount of nonperforming worth with another pro-cyclical feedback tability and further deteriorated their net Overall, the initial shock to the fi into the aggregate economic activity. intensifi sector itself seems to have caused and/or and through the fi worth of all three groups of economic agents effect generated a severe crisis. that are sensitive to the free fl that are sensitive to debt capacity further increased external fi debt capacity further increased external The fi planation of the proximate causes of the recent crisis. If we consider the collapse the If we consider crisis. of the recent causes proximate of the planation economic of 2007 as the initial in September mortgages market of the subprime the size of as large. Although be categorised this shock can hardly shock, then it was less terms ($0.7 trillion) large in absolute mortgages market was subprime of US fi of the size than 0.5 percent source of the could become the on these loans is puzzling how default Hence, it be summarised as fol- this era. A possible explanation can most severe crisis of fi the US subprime mortgage market forced lows. The collapse of caused an erosion of billion dollars in bad loans and diaries to write off hundreds sheets harmed their Deterioration of intermediaries’ balance their capitalization. Consequently, lending which reduced their lending capacity. ability to raise funds, standards that non-fi and property prices affected investment, consumption The credit crunch adversely causes of the recent are still a matter of debate and intensive research. At the moment a few causes of the recent financial crisis are still a matter of debate and intensive research. What was the source questions seem to be crucial with respect to the ultimate causes of recent financial crisis. of excessive liquidity on the financial markets of the US and other industrialized countries in the years before what extent did financial innovations and financial derivatives, which To outbreak of recent financial crisis? amp- had been supposed to make the financial system more stable by diversifying risk and rising liquidity, Why did financial institutions engage in excessive risk taking? lify the problem of information asymmetry? what extent might the principal-agent problem between the owners and managers of financial institutions To contribute to excessive risk taking? ties of many other countries as well. Using the dynamic general equilibrium mo- ca- po- cation equiva- cance of this effect ows can be genera- nancial accelerator effect cult is that the existence of a et nancial accelerator (Bernanke condition cient but not a necessary nance premium. nance of Ger- The results nance premium. rst-principle based macroeconomic rst-principle based macroeconomic activity during the aggregate economic cial effect on accelerator effect. nancial suffi nance premium is a culties. Problems that are commonly acknowledged are: culties. Problems that are commonly net in banks’ to changes emerges due effect accelerator nancial nancial accelerator effect even more diffi a positive correlation between aggregate credit ows, that is, the existence of nancial accelerator effect suggests that adverse economic disturbances lead nancial accelerator ows are caused solely by shifts in investment demand. The inability to distin- recent popularity in the broader econo- accelerator. However, despite its nancial ts from these credit market measures increase with the severity of the crises, with the severity measures increase these credit market ts from al., 1996). This testing procedure suffers from the so-called identifi al., 1996). This testing procedure suffers between aggregate credit changes and lence problem. The positive correlation GDP changes can be caused not just by movements in credit supply but by move- ments in credit (investment) demand as well. For example, King and Plosser (1984) demonstrated that procyclical movements of aggregate credit fl ted by a frictionless real business cycle model, in which changes in aggregate cre- dits fl guish between different sources of observed colinearity makes this testing proce- dure unsuitable. Hence, the empirical literature has followed alternative identifi tion strategies. These empirical investigations consist of two groups of studies. and aggregate output movements. The strong positive correlation between these and aggregate output movements. The time series (see for example: Del- variables is, in general, observed in aggregate the joint test based on aggregate data is l’Ariccia and Garibaldi, 2005). However, of fi unlikely to be informative about the existence to a decrease in aggregate investment due to a shift in the credits supply curve to a decrease in aggregate investment costs. Hence, it seems that empiri- caused by increases in asymmetric information gate of procyclical movements in aggre cal tests can focus on the possible existence credit fl is still relatively weak. The empirical estimates of the fi is still relatively weak. face three main diffi The fi Theoretical consistency and the ability to offer a plausible explanation of the recent and the ability to offer a plausible explanation Theoretical consistency toward the idea of the of a broader economic community crisis has drawn attention fi empirical evidence on the economic signifi mic community, the del in which the fi which the del in the net bene- demonstrated that simulations also (2010) model tler and Kiyotaki’s fi in the cases employ them only it makes sense to to account for why which helps they demonstrate that These results are important because of severe recession. in line with modern, fi these measures are theory. worth Gertler and Kiyotaki (2010) demonstrated that implementation of this kind of this implementation that demonstrated (2010) Kiyotaki Gertler and worth can have benefi of measures in the external fi reducing the rise the crisis by for the existence of the fi procyclical external fi tential endogeneity that hampers any evaluation of aggregate data, and inability tential endogeneity that hampers any evaluation fi directly to observe and measure the external and that makes an empirical assessment A problem that has not been recognised, of the fi

FINANCIAL THEORY AND BRUNO ĆORIĆ: PRACTICE THE FINANCIAL ACCELERATOR EFFECT: CONCEPT AND CHALLENGES 35 (2) 171-196 (2011) 186 FINANCIAL THEORY AND BRUNO ĆORIĆ: PRACTICE THE FINANCIAL ACCELERATOR EFFECT: CONCEPT AND CHALLENGES

187 35 (2) 171-196 (2011) r- ca- nance nance should 12 rms that rms nance pre- rms are very es the effects of es the effects nancial accelerator nancial accelerator, a nancial accelerator, nancial markets. During nancial be supposed rms that can cant recession in the cases of since fi nance premium lar- rms are on average considerably In particular, nancial accelerator effect. nancial accelerator effect is based on a effect is based nancial accelerator rmed by Mody and Taylor (2004), who rmed by Mody fi cation of a countercyclical external ect investors belief in the “too big to fail” infor- that are more exposed to asymmetric rms nance premium and aggregate economic activity. activity. economic and aggregate premium nance nancial accelerator effect in the banking sector. Al- nancial accelerator effect in the banking sector. found a accelerator theory. In particular, they nancial rms increase less during recession than bond rates on rms increase less during recession nance premium that is not directly observable. Gertler observable. Gertler that is not directly nance premium nance (bond rate), according to the fi rms. Second, identifi rms that raise funds on the high-yield bond market. The recent econo- rms that raise funds on the high-yield GDP. There- ed, which can make them less exposed to change in domestic rst group of studies focuses on the relationship between variables that are that variables between the relationship on studies focuses group of rst The high-yield bond spread is the difference between high-yield bond rate and the corresponding rate for The high-yield bond spread is the difference the bonds of AAA rated firms. AAA the bonds of 12 aggregate shocks. Third, the results of these studies are based only on US data, aggregate shocks. Third, the results of these studies are based only on US data, using a relatively short series, with only one signifi Gertler and Lown (1999) and Mody and Taylor (2004). Fourth, due to the short time period and lack of data for other countries, the potentials for further applica- tion of this approach are limited. Finally, the existence of external fi mium is not a necessary condition for the fi tion of AAA rated fi this premium intensifi premium does not necessary imply that countercyclical external fi countercyclical market (the of a high-yield debt the development (1999) argued that and Lown resolve this US can potentially early 1980s in the market) in the “junk bonds” high-yield bond spread and Lown (1999) a to Gertler problem. According The fi fi the external to proxy meant That is, in the majority of models, the fi the majority of models, That is, in fore, the negative correlation between high-yield spread and annual percentage fore, the negative correlation between of the larger international diversifi changes in real GDP can also be the results the high-yield bond market can refl are usually more internationally di- hypothesis. Furthermore, the large companies versifi mic crisis has demonstrated that size (still) matters. Hence, the evidence that the mic crisis has demonstrated that size (still) bond rates of AAA rated fi raise funds by issuing high-yield bonds are the kind of fi raise funds by issuing rated fi credit markets. On the other hand, AAA to face frictions on information problems on fi unlikely to face asymmetric fi recession the cost of more for fi effect, should increase reverse should be true high-yield bond spread should rise. The mation. Hence, the empirical tests confi phase. The results of their over the economy expansionary of the fi med this prediction output gap. The results between high-yield spread and strong inverse relationship of Gertler and Lown (1999) were confi high-yield spread and annual per- recorded a high negative correlation between Aliaga-Diaz and Olivero (2010) found centage changes in real GDP. Recently, also consistently countercyclical, which that price-cost margins for US banks are can be an indicator of the fi the theory of the fi though these results are consistent with some concerns about this testing ap- few issues challenge robustness and raise spread is good proxy for external fi proach. First, whether high-yield bond fi premium is a matter of debate; AAA rated ger than fi provide a good overall indicator of the external fi provide a good overall

- - . If na- 13 rms’ cation. rm and However, 14 rms’ real in- nancial accelerator nance (net worth) imper- nancial market fi ndings that external nancial constraints, i.e. which rmed by large number of empirical stu- variable that measures fi cient on the nancial accelerator effect. For example, the rms’ investment and banks’ lending. In brief, rms’ investment and banks’ lending. nancial (equity) markets can amplify and propa- can amplify markets (equity) nancial of fi nance are an important determinant rms more likely to face fi should be cient and increase of the explanatory power nance on fi be positive and statistically signifi nance (net worth) should ndings have been confi rms’ investment. Underlying the FHP approach is the premise that rms’ investment. Underlying the FHP approach is the premise In particular, also has cross sectional implications. nancial accelerator theory uence on economic agents that face severe asymmetric information problems agents that face severe asymmetric information uence on economic Firms’ internal source of finance is usually proxied by cash flow. Firms’ can also be found Empirical evidence consistent with some implications of the financial accelerator effect rms’ ability and/or the terms under which they are able to borrow, hence also rms’ ability and/or the terms under which they are able to borrow, hence have a stronger predicts that economic shocks should nancial accelerator theory of this literature, hence, this nancial accelerator effect is not in the primary focus in: Peersman and Smets (2005), Almeida et al. (2006), Mody et al. (2007) and Cavalcati (2010). in: Peersman and Smets (2005), 13 14 larger for the categories of fi the information asymmetry on fi asymmetry the information worth agents’ net economic in changes procyclical shocks because gate economic to risk (see agents’ exposure in economic change produce countercyclical fi In that respect, possible and Stigliz, 1993). Greenwald exist (that does not even or that a premium is not countercyclical nce premium that the fi would not imply are frictionless) credit markets (fi effect. This literature employs the disaggregated nancial accelerator the effects of change in for the cross sectional difference in bank level) data to test the internal source of fi Hubbard and Petersen (1988) (FHP). FHP the literature was initiated by Fazzari, for the effect of fi used Tobin’s Q investment model to test fections on fi fi internal sources of fi their investment, are a function of their effect is irrelevant. effect is irrelevant. movements over time, prediction about aggregate investment As well as being a the fi fi infl on the role of infor- agents. The numerous panel data studies than on less exposed can be considered as determining real investment decisions mation asymmetry in in support of the fi studies that provide empirical evidence the second group of due to some serious shortcomings of this testing approach, these results cannot be due to some serious shortcomings of this testing approach, these results cannot considered robust evidence of the fi fi this is the case then the estimated coeffi internal sources of fi equation. Moreover, its inclusion cant when it enters the investment regression of the standard investment specifi should increase the explanatory power Finally, the estimated coeffi problem. The results of FHP are more exposed to the asymmetric information these predictions supporting the hypothe- empirical estimations were in line with sis that internal sources of fi vestments. These fi dies (for references and a recent review of this literature see Ćorić, 2010). Essen- dies (for references and a recent review has been used to test the role of in- tially the same approach, with similar results, lending decisions, that is, to test the formation asymmetry in determining banks’ credit channel of monetary policy (see for example: Kashyap and Stein, 1995, 2000; Kishan and Opiela, 2000; and Gambacorta and Mistrulli, 2004).

FINANCIAL THEORY AND BRUNO ĆORIĆ: PRACTICE THE FINANCIAL ACCELERATOR EFFECT: CONCEPT AND CHALLENGES 35 (2) 171-196 (2011) 188 FINANCIAL THEORY AND BRUNO ĆORIĆ: PRACTICE THE FINANCIAL ACCELERATOR EFFECT: CONCEPT AND CHALLENGES

189 35 (2) 171-196 (2011) - nan- nancing ow coef- ow nancial accelerator ef- of these models. cance nancing constraints. Third, nancing constraints. rms’ net worth is procyclical and and is procyclical net worth rms’ uctuations depends then on the na- the recent are. Furthermore, ndings in the opposite direction, ation works nancing constraints proposed by Ho- nancing constraints proposed es any reduction in borrowers’ net worth, cation of fi much smal- disappears or becomes ow effect either Second, the recent literature demonstrates that the fi Second, the recent literature demonstrates ation amplifi ation 15 nancial accelerator effect. Conversely, in the case of critique that the ex- nance premium it is not subject to the dence. cients can be generated without any fi be generated without cients can cash-fl constraints exist, the size of the estimated nancing the existence of the fi is not a necessary condition for nance premium ow coeffi nancial accelerator effect predicts that this effect should be stronger the nancial accelerator effect predicts that this effect Whether the analysed effect is going to work as accele- ation effect emerges. The rationale for expansive short-run during is also to a large extent based on the The rationale for expansive short-run fiscal policy during recessions is also to a large which under- related to the degree of the constraints, cient need not be positively assumption that in recessions (especially deep recessions) government spending should act to substitute for assumption that in recessions (especially deep recessions) government spending should act to substitute lower demand for private investments (see for example, recommendations for fiscal policy during the recent et al., 2009). crisis in Spilimbergo 15 what the potential aggregate effects of these fi of these effects aggregate the potential what for fi on the evidence that cast doubt produced three results research has literature does not assess to what extent the fi extent the to what not assess does literature ler when one controls for the measurement error in Tobin’s Q. Second, positive in Tobin’s Q. Second, measurement error controls for the ler when one cash-fl constraints from the studies based on FHP methodology (Cummins, Hassett and (Cummins, on FHP methodology from the studies based constraints First, the cash-fl Oliner, 2006). nancial accelerator effect theoretically can go in the opposite direction, “decelera- nancial accelerator effect theoretically can volatility (see: Iacoviello, 2005; Hou- ting” economic shocks and reducing output Iacoviello (2005), for example, de- se, 2006; and Christiansen and Dib, 2008). contracts are written in the terms of the monstrated that in a case in which debt the additional possibility of Fisher’s debt- nominal interest rate, as is usually done, defl fl rator or rather as “decelerator” of economic de- ture of the economic shocks. In particular, in the case of a negative aggregate mand shock, decline in infl and due to an increase in the real value of borrowers’ outstanding debt obligations, hence, exaggerates the fi adverse aggregate a rise in infl ternal fi assuming that fi is less subject to (2006) and Almeida and Campello (2007) vakimian and Titman does not test for the Second, since this estimation procedure the above critiques. fi existence of external fi (for a detailed analysis assumption of FHP methodology cuts the key identifying this literature does see Ćorić, 2010). Although at the moment of these researches of research. First, the this is a potentially promising line not provide robust results, for identifi recent methodology deeper the recession is (due to the lower level of net worth). This implication deeper the recession is (due to the lower business and economic community (see seems to contradict common belief in the the deepest stages of recession the low for example, Chatterjeee, 2010), that in demand caused by great uncertainty and credit level is caused by lack of credit low business confi fect, more research is needed to assess the empirical signifi fect, more research is needed to assess is especially urgent, for two reasons. The necessity of more empirical researches The fi Overall, despite ample theoretical research based on the fi Overall, despite ample theoretical research cial accelerator effect. Finally, data for this kind of empirical analysis are relati- cial accelerator effect. Finally, data for vely available, at least for developed countries. rm or nancial accelerator nancial accelerator fra- nancial accelerator models ed when banks are lenders. Howe- nancial accelerator effect would still exist effect. nancial accelerator cial for the borrowers’ net worth and conse- worth and net the borrowers’ cial for are not the ma- rms, especially in banks, the owners ed when banks are borrowers. Although banks are forced ed when banks are borrowers. Although are informed about their net rms, and especially households rm’s or bank’s net worth. Therefore when the principal-agent nancial accelerator effect and its introduction in the fi nancial accelerator effect and in this case since a higher economic agent net worth also implies a lower probabi- in this case since a higher economic agent that in this case it is relative rather than lity of bankruptcy, and also to speculate to a lender. However, this should be absolute net worth that would be important investigated in more detail. individual economic agent’s net worth is The literature also presumes that only an a lender is able to observe a borrower’s important. The underlying premise is that justifi net worth. This assumption can be easily ver, it seems less justifi on a regular basis it still remains to release information on their capitalization doubtful how much fi do have information about these data re- worth. For example, even if households have enough knowledge to understand leases it is very doubtful whether they different banks. Consequently, it would them and compare capitalization among of banks, information about soundness of be useful to analyse whether in the case the overall banking sector that can be communicated through the media or central is banks’ announcements; what the function of banks’ aggregate net worth can be also important. With respect to the results of existing fi problem exists, an economic agent’s net worth can lose its relevance for the len- problem exists, an economic agent’s net der. It is possible to speculate that the fi reducing real value of borrowers’ outstanding debt obligations. In that respect In that obligations. debt outstanding of borrowers’ real value reducing be benefi shocks can supply adverse shock. an economic quently “decelerate” new aspects we suggest two theoretical research, to possible further With respect of the fi the existing concept to expand asymmetry between for the problem of information Existing literature accounts “within economic age- not for the information asymmetry economic agents, but many fi nts”. In particular, in net worth is important Namely, an economic agent’s mework can be informative. agent will behave high net worth implies that an economic for lenders because of bankruptcy will take less risk since its losses in the case more diligently and/or owners of a fi in a case in which managers are not the be larger. However, (loss of reputation) seem to be inde- bank, their losses in the case of bankruptcy pendent of the fi on this would not make difference since their mathematical formulation is based ex- the representative agent approach. However, this might be important for the planation and especially for the empirical assessment of the fi nagers. Asymmetric information between owners and managers results in what is information between owners and managers nagers. Asymmetric important for the problem. This problem is potentially called the principal-agent fi effect.

FINANCIAL THEORY AND BRUNO ĆORIĆ: PRACTICE THE FINANCIAL ACCELERATOR EFFECT: CONCEPT AND CHALLENGES 35 (2) 171-196 (2011) 190 FINANCIAL THEORY AND BRUNO ĆORIĆ: PRACTICE THE FINANCIAL ACCELERATOR EFFECT: CONCEPT AND CHALLENGES

191 35 (2) 171-196 (2011) nan- nancial nancial nancial cri- nancial rst-principle rst-principle nancial markets and short-run nancial markets and short-run nancial markets and short-run markets and nancial effect. nancial accelerator scal authorities. scal cation of this effect. Overall, we found cation of this effect. Overall, nancial accelerator effect. More empiri- nancial accelerator framework; and also literature has faced se- cance, the empirical nancial accelerator effect recently very popular in the nancial accelerator effect recently rm its signifi uctuations based on informational asymmetry on fi uctuations based fi the so-called uctuations, nancial accelerator effect offers a consistent, fi offers a consistent, accelerator effect nancial CONCLUSION

markets. This effect also offers a plausible rationalization of the severe conse- also offers a plausible rationalization markets. This effect of 2007. Finally, mortgages market’s crash in September quences of the subprime manifestation, precisely, the prevention of its even stronger this effect, or more taken during the background for the credit policy measures provides a theoretical central banks and fi recent crisis by many the fi These features made the recent fi Despite its popularity, and broader economic community. sis, which seems to confi rious challenges in the empirical identifi unable to provide robust assessments of that existing empirical literature is still fi the size and economic relevance of the the recent literature demonstrates that cal research is necessary especially because opposite direction and reduce output vola- this effect theoretically could go in the two new lines of enquiry. In particular, tility as well. Finally, we made a case for to introduce the principal-agent problem we found that it might be informative fi between owners and managers into the worth of banking sector in the fi to consider the role of the aggregate net accelerator effect and its empirical estimation. based, explanation of the relationship between fi based, explanation fl aggregate economic We found that the fi We found 6 of fi the role about a question crisis raises economic the recent of The severity cial markets in modern market economies. This review concentrates on the rela- review concentrates economies. This in modern market cial markets asymmetry on fi information tionship between economic fl aggregate , 9 The Econo- , 82 (4), , 128 (1), The Quar- , 73 (3), 257- , 20 (5), 1429- Brooking Papers Brooking , 8 (1), 163-178. , 10 (3), 321-352. Journal of Finan- nancial accelerator nancial , No. 11349. , No. 11349. in US accelerator nancial “The fi No. 22. London: Bank of nancial accelerator: Eviden- , (1), 87-114. , 79 (1), 14-31. “House Prices, Consumption, “House Prices, Consumption, “Houses as Collateral: has the “Houses as Collateral: New York: Oxford University “The fi “The Federal Funds Rate and the Review of Finance “Is it Money or Credit, or Both, or “Is there a fi “Is there a Journal of Economic Perspectives American Economic Review , 78 (2), 435-439. “Agency Costs, Net Worth, and Busi- “The Credit Crunch”. “Financial Fragility and Economic Per- “Inside the Black Box: The Credit Chan- American Economic Review , 44 (1), 48-58. “Financial Constraints, Asset Tangibility, “Financial Constraints, “Measuring Monetary Policy”. “How Much does Bank Capital Matter?”. does Bank Capital “How Much NBER Working Paper NBER Working The Review of Financial Studies The Review of Financial The Review of Economics and Statistics , 113 (3), 869-902. Financial Stability Report,

“Volatility and Growth: Credit Constraints and Produc- Credit Constraints Growth: and “Volatility , 2, 205-247. Business Cycle Theory. “Non-Monetary Effect of the Financial Crisis in the Pro- “Non-Monetary Effect of the Financial American Economic Review

, 13 (4), 414-435. .], 2005. , 108 (2), 167-171. The Quarterly Journal of Economics AL American Economic Review

ET , P. [ ce from International Housing Markets”. ce from International Link between House Prices and Consumption in the U.K. Changed?”. Prices and Consumption in the U.K. Link between House Review of New York Federal Reserve Bank Approach”. and Monetary Policy: A Financial Accelerator cial Intermediation formance”. Press. ness Fluctuations”. nel of Monetary policy Transmission”. England. pagation of the Great Depression”. 276. Neither?”. Channels of Monetary Transmission”. 901-921. (4), 27-48. 1-15. and the Flight to Quality”. on Economic Activity on Economic terly Journal of Economics tivity-Enhancing Investment”. Investment”. tivity-Enhancing Quarterly Bulletin Bank of England 1460. banking? Evidence from the cyclicality of banks’ price-cost margins”. price-cost margins”. of banks’ Evidence from the cyclicality banking? mics Letters and Corporate Investment”. GHION Almeida, H., Campello, M. and Liu, C., 2006. Almeida, H., Campello, J. and Vlieghe, G., 2002. Aoki, K., Proudman, J. and Vlieghe, G., 2004. Aoki, K., Proudman, LITERATURE A Arnold, L. G., 2002. Bernanke, B. S. and Gertler, M., 1990. Bank of England, 2007. Bernanke, B. S. and Gertler, M., 1995. Bernanke, B. S., 1983. 1988. Bernanke, B. S. and Blinder, A. S., 1992. Bernanke, B. S. and Blinder, A. S., Bernanke, B. S. and Gertler, M., 1989. Bernanke, B. S. and Lown, C. S., 1991. Bernanke, B. S. and Mihov, I., 1998. Bernanke, B. S., Gertler, M. and Girlchrist, S., 1996. Almeida, H. and Campello, M., 2007. Almeida, H. and Campello, Aikman, D. and Vileghe, G., 2004. D. and Vileghe, G., Aikman, P. M., 2010. R. and Olivero, Aliaga-Diaz,

FINANCIAL THEORY AND BRUNO ĆORIĆ: PRACTICE THE FINANCIAL ACCELERATOR EFFECT: CONCEPT AND CHALLENGES 35 (2) 171-196 (2011) 192 FINANCIAL THEORY AND BRUNO ĆORIĆ: PRACTICE THE FINANCIAL ACCELERATOR EFFECT: CONCEPT AND CHALLENGES

193 35 (2) 171-196 (2011) . cation American Journal of , 53 (2), 219- Handbook of Handbook Review of Eco- NBER Working NBER Working Oxford Bulletin of Bulletin Oxford “Investment Beha- nancial accelerator accelerator nancial , 11 (1), 155-178. American Economic Re- “The fi “The , 1, 135-174. , 20 (3), 559-597. “An Estimated Small Open , 34 (4), 407-434. , 13 (4), 436-457. International Economic Re- International , 2Q, 1-8. “Balance Sheets and Exchan- “Balance Sheets and IMF Staff Papers nancial accelerator: How Wall nancial accelerator: scal stimulus during the crisis”. , 94 (4), 1183-1193. “Does Bank Capital Affect Lending nancial accelerator in an estimated , 75 (4), 850-855. “Agency Cost, Net Worth, and Busi- “Agency Cost, Net “Gross Credit Flows”. “Collateral and Rationing: Sorting Equi- and Rationing: Sorting “Collateral accelerator, Globalization and Ou- nancial “The Long and large Decline in US Output “The Long and large “The fi , 57 (5), 506-526. Business Review “The fi “Credit Market Imperfection and the Power of the “Credit Market Imperfection “Global effects of fi Review of Economic Dynamics (forthcoming). nancial accelerator”. Social Science Research Network Working Paper Network Working Social Science Research “Net Worth, Exchange Rates, and Monetary Policy: The . Amsterdam: North-Holland, 1341-1393. North-Holland, . Amsterdam: Financial Theory and Practice “De-Laveraging and the fi 83 (5), 893-910. , 32 (1), 118-144. American Economic Review “Financial Structure and Aggregate Economic Activity: An “Screening vs. Rationing in Credit Markets with Imperfect In- Credit Markets with vs. Rationing in “Screening , 72 (3), 665-685. Journal of Financial Intermediation “Investments and Capital Markets Imperfections, Identifi “Investments and Capital Markets Imperfections, “The Global Extent of the Great Moderation”. American Economic Review American Economic Journal of Money, Credit and Banking Credit Journal of Money, Brooking Papers on Economic Activity Papers on Economic Brooking , No. 13244. , 96 (3), 796-810. , 28 (3), 671-689. nancial accelerator: A Theoretical and Empirical Investigation”. A Theoretical and Empirical nancial accelerator: Paper Overview”. Journal of Behaviour?”. Effects of Devaluation in a Financially Fragile Environment”. 241. nomic Studies Economy Model of the fi ge Rate Policy”. Street Can Shock Main Street”. New Keynesian model”. Funds”. viour, Observable Expectations, and Internal Economic Review, Issues: A Survey”. tput Growth Volatility”. view ness Fluctuations: A Computable General Equilibrium Analysis”. Computable General Equilibrium Analysis”. ness Fluctuations: A Economics and Statistics formation”. Volatility”. fi in a Quantitative Business Cycle Framework” in J. B. Taylor, ed. in J. B. Cycle Framework” Business in a Quantitative 1 Macroeconomics Markets”. and Competitive libria in Monopolistic view Gertler, M., 1988. Gatti, D. [et al.], 2007. Freedman, C. [et al.], 2010. Gambacorta, L. and Mistrulli, P. E., 2004. Dell’Ariccia, G. and Garibaldi, P., 2005. Elekdag, S., Justiniano, A. and Tchakarov, I., 2006. Elekdag, S., Justiniano, A. and Tchakarov, Cespedes, L. F., Chang, R. and Velasco, A., 2004. Cespedes, L. F., Chang, Chatterjee, S., 2010. Christensen, I. and Dib, A., 2008. Oliner, S. D., 2006. Cummnis, J. G., Hassett, K. A. and Carlstrom, T. C. and Fuerst, T. S., 1997. Carlstrom, T. C. and Ćorić, B., 2010. Ćorić, B., 2011. Ćorić, B. and Pugh, G., 2011. Bester, H., 1985. Bester, H., Blanchard, O. and Simon, J., 2001. Blanchard, O. and H., 2010. Calvalcanti, M. F. Besanko, D. and Thakor, V. A., 1987. D. and Thakor, V. Besanko, Bernanke, B. S., Gertler, M. and Girlchrist, S., 1999. Girlchrist, M. and Gertler, B. S., Bernanke, Fe- , 40 , 42, Scan- (forthco- Journal of Oxford Re- Oxford American Eco- American Econo- , 112 (3), 663-691. , 108 (1), 77-114. , 15 (4), 432-444. , 95 (3), 739-764. Amsterdam: Elsevier “Are There Bank Effect “Informational Imperfec- , 75 (4), 559-581. “External Constraints on Mo- “External Constraints Journal of Money, Credit, and Credit, Journal of Money, nancial accelerator”. nancial “Financial Market Imperfections and “Financial Market , 77 (3), 63-77. , 38 (2), 357-374. “Corporate Investment and Financial “Corporate Investment and Financial “What Do a Million Observations on “The impact of Monetary Policy on Bank Journal of Money, Credit and Banking Credit Journal of Money, “The Information in the High-Yield Bond in the High-Yield “The Information “The Role of Credit Market Imperfections Imperfections Credit Market Role of “The “Financial Intermediation, Loanable Funds, “Financial Intermediation, Loanable Funds, “Financial Intermediation and Credit Policy Intermediation and “Financial “Measuring and Comparing Business-Cycle “Measuring and Comparing Business-Cycle Handbook of Monetary Economics of Monetary Economics Handbook , 95 (1), 43-64. Journal of Business “Credit Rationing”, in: B. M. Friedman and F. “Credit Rationing”, in: B. M. Friedman American Economic Review , 15 (3), 132-150. nancial accelerator”. The Quarterly Journal of Economics , 53 (6), 1117-1134. “Is There a ‘Credit Channel’ for Monetary Policy?”. “Is There a ‘Credit Channel’ for Monetary The Quarterly Journal of Economics The Quarterly Journal “House Prices, Borrowing Constraints, and Monetary Poli- “House Prices, Borrowing Constraints, Carnegie-Rochester Conference Series on Public Policy Carnegie-Rochester Conference “Adverse selection and the fi Handbook of Monetary Economics 2. “Equilibrium Indeterminacy and Asset Price Fluctuations in “Equilibrium Indeterminacy and Asset , 74 (2), 194-199. , 90 (3), 407-428. Journal of Business and , 39 (2-3), 295-330. . tions in the Capital Market and Macroeconomic Fluctuations.” and Macroeconomic Fluctuations.” tions in the Capital Market 151-195. Banks Say About the Transmission of Monetary Policy?”. mic Review in Business Cycle Analysis” in Cycle Analysis” in Business fi netary Policy and the Banking nomic Review cy in the Business Cycle”. Science Publishers, 837-888. Balance Sheets”. Features”. in the Monetary Transmission Mechanism: Arguments and Evidence”. and Evidence”. Arguments Mechanism: Transmission Monetary in the of Economics dinavian Journal ming) Cycle: Evidence and Some Implications”. Spread for the Business view of Economic Policy Business Cycles”. (5), 967-999. and the Real Sector”. in Borrowers Cost of Funds?”. Japan: A Bayesian Investigation”. H. Hahn, eds. Monetary Economics deral Reserve Bank of St. Louis Review Constraints: Sensitivity of Investment to Funds from Voluntary Asset Sales”. Constraints: Sensitivity of Investment to and Banking Credit, Journal of Money, Kashyap, A. K. and Stein, J. C., 2000. Gertler, M. and Kiyotaki, N., 2010. and Kiyotaki, N., Gertler, M. Stiglitz, E. J., 1993. Greenwald, B. C. and Hess, G. D. and Iwata, S., 1997. Iacoviello, M., 2005. Kashyap, A. K. and Stein, J. C., 1995. Gertler, M., Gilchrist, S. and Natalucci, F., 2007. Gertler, M., Gilchrist, Gertler, M. and Gilchrist, S., 1993. Gilchrist, M. and Gertler, 1999. and Lown, C. S., Gertler, M. E. J., and Weiss, A., 1984. Greenwald, B., Stiglitz, Hirose, Y., 2008. Holmstrom, B. and Tirole, J., 1997. House, C. L., 2006. D. N., 2002. Hubbard, R. G., Kuttner, K. N. and Palia, Jaffe, D. and Stiglitz, E. J., 1990. Hubbard, R. G., 1995. Hovakimian, G. and Titman, S., 2006.

FINANCIAL THEORY AND BRUNO ĆORIĆ: PRACTICE THE FINANCIAL ACCELERATOR EFFECT: CONCEPT AND CHALLENGES 35 (2) 171-196 (2011) 194 FINANCIAL THEORY AND BRUNO ĆORIĆ: PRACTICE THE FINANCIAL ACCELERATOR EFFECT: CONCEPT AND CHALLENGES

195 35 (2) 171-196 (2011)

, 9, , 10 Journal , 8 (1), 259- nancial acce- cation through through cation CESifo Forum , 32 (1), 121-141. , 21 (2), 265-293. Journal of International Journal of International Journal of Political Eco- Journal of The Japanese Economic Re- The Japanese Journal of International Eco- “A cross-country fi , 71 (3), 393-410. , 56 (2), 242-254. , 82 (2), 167-172. , 56 (4), 596-604. “Corporate Financing and Investment “Corporate Financing and Investment “On Balance Sheets, Idiosyncratic Risk “On Balance Sheets, Idiosyncratic Risk , 32 (1), 68-89. “Bank Size, Bank Capital, and the Bank Bank Capital, and “Bank Size, , 74 (3), 363-380. 363-380. , 74 (3), “Financial shocks and the US business “Financial shocks and the US business , 115 (503), 319-342. , 7 (4), 1-36. “Money, Credit, and Prices in Real Busines Busines in Real and Prices Credit, “Money, “Credit Cycles”. “Credit Cycles”. “The Industry Effects of Monetary Policy in “The Industry Effects of Monetary Policy “Theoretical notes on bubbles and the current “Theoretical notes on “Credit Rationing in Markets with Imperfect “Financial Predictors of Real Activity and the “Financial Predictors , No. 16399. “Fiscal Policy for the Crisis“. The B. E. Journal of Macroeconomics (topics) The B. E. Journal of Macroeconomics Journal of Economic Theory “Does Capital Matter for Monetary Transmission?”. , 13 (2), 187-221. Economic Letters “Optimal Contracts and Competitive Markets with Co- , 26 (1), 149-165. Journal of Money Credit and Banking Money Credit Journal of “Aggregate Gains of International Diversifi “ mismatch, openness and exchange rate regime openness and exchange rate regime “Currency mismatch, “ in banking, countercyclical margins and the “Market power in banking, countercyclical “New Keynesian models, durable goods, and collateral con- “New Keynesian models, durable goods, cation”. “Credit and Business Cycles”. Business Cycles”. “Credit and Economic Journal American Economic Review Journal of Monetary Economics Global Economy Journal Journal of Macroeconomics American Economic Review Economic American NBER Working paper NBER Working Journal of Monetary Economics , 80 (2), 292-301. , 105 (2), 211-248. , 49 (1), 18-35. nancial accelerator”. accelerator”. nancial The Federal Reserve Bank of New York Economic Policy Review The Federal Reserve Bank of New York 266. stly State of Verifi Information”. (2), 26-32. Foreign Direct Investment: An Inquiry into the Moderation of U.S. Business Foreign Direct Investment: An Inquiry Cycles”. and Aggregate Volatility”. the Euro Area”. Article 4. Decisions when Firms have Information that Investors do not have”. Decisions when Firms have Information of Money and Finance straints”. cycle”. lerator: Evidence from North America and Europe”. lerator: Evidence from nomics international transmission of business cycles”. international transmission of business crisis”. fi choice”. nomy Cycle”. Cycle”. Lending Channel”. Lending Channel”. view Van den Heuvel, J. S., 2002. Townsend, R. M., 1979. Stiglitz, E. J. and Weiss, A., 1981. Spilimbergo, A. [et al.], 2009. Portes, L. S. V. and Ozenbas, D., 2009. Peersman, G. and Smets, F., 2005. Portes, L. S. V., 2007. Nolan, C. and Thoenissen, C., 2009. Monacelli, T., 2009. Myers, C. S. and Majluf, S. N., 1984. Mody, A., Sarno, L. and Taylor, M. P., 2007. Mody, A., Sarno, L. Olivero, P. M., 2010. Martin, A. and Ventura, J., 2010. Martin, A. and Ventura, M. P., 2004. Mody, A. and Taylor, Magud, E. N., 2010. Kiyotaki, N. and Moore, J., 1997. N. and Moore, J., Kiyotaki, King, R. G. and Plosser, C. I., 1984. Plosser, R. G. and King, Kishan, R. P. and Opiela, T. P., 2000. P. and Opiela, T. Kishan, R. Kiyotaki, N., 1998. Kiyotaki, , 111 (3), , 64 (3), 213- nancial acce- nancial , 94 (4), 944-963. “The Macroeconomics of Labour and Credit “The Macroeconomics The Scandinavian Journal of Economics Journal of The Scandinavian “The Bank Capital Channel of Monetary Policy”. Policy”. of Monetary Channel Bank Capital “The “How important are Financial Frictions in the United Frictions in the United are Financial “How important American Economic Review American Economic Oxford Bulletin of Economics and Statistics Bulletin of Economics Oxford “Business Fixed Investment: Evidence of a fi Fixed Investment: “Business University of Pennsylvania. University

, Working paper Working lerator in Europe”. lerator in 231. States and the Euro Area?”. States and 567-596. Market Imperfections”. Market Imperfections”. Van den Heuvel, J. S., 2009. J. S., den Heuvel, Van P., 2002. Vermeulen, von Heideken, V. Q., 2009. von Heideken, Wasmer, E. and Weil, P., 2004. Wasmer, E. and Weil,

FINANCIAL THEORY AND BRUNO ĆORIĆ: PRACTICE THE FINANCIAL ACCELERATOR EFFECT: CONCEPT AND CHALLENGES 35 (2) 171-196 (2011) 196