Karen Mills, Steven Morling and Warren Tease
May 1994
- Download the Paper53KB
Financial factors play a limited role in traditional models of investment. For example,in the neoclassical model, firms choose inputs of capital (and labour) so asto maximise the present discounted value of their income streams. Financialfactors enter only through the cost of capital which, in turn, is independentof the way the firm finances itself. This independence arises because capitalmarkets are assumed to be perfect. Thus, firms are able to secure externalfinance for a project if its expected marginal return exceeds its cost of capital.There is no shortage of funds for firms with value-increasing projects andthe marginal costs of debt, equity and internal funds are equal. In this world,the availability of adequate cash flows is not a constraint on investment andthe financial characteristics of the firm do not influence its cost of capital.Thus, “interactions between real and financial variables can be reducedto interactions between real variables and interest rates” (Mauskopf(1990)).
There are a number of reasons to believe that this separation of real and financialfactors would not occur in practice. Some firms (particularly small firms)have limited access to external sources offunding.[3]Cash flows will be their primary, and in some cases, only source of funds.
Even companies with access to external funding will rely more heavily on cash flowsas a source of finance. There are direct costs involved in raising externalfunding, such as underwriting and administrativefees.[4]There are also potential financial distress costs associated with using externalfinance. For example, as leverage increases, other things being equal, theremay be a higher probability of the firm facing financial distress. In thiscase, the firm may incur direct bankruptcy costs such as legal expenses andtrustee fees and indirect costs such as the disruption of operations, lossof suppliers or customers and the imposition of financial constraints. Thepresent value of these expected costs should be reflected in current financingcosts (whether or not the firm actually enters bankruptcy). Finally, thereare issues of taxation, shareholder dilution, control of information, the needto maintain flexibility and liquidity that may also have an impact on a firm'sfinancing choices. Financial factors may therefore affect the cost and availabilityof capital and so influence the investment decision.
Financial factors are generally introduced to standard investment models throughinformation asymmetries or through agency costs. The introduction of theseassumptions helps explain how a given level of investment will be funded andhow a firm's financial position will influence its investment.
Informational asymmetries, where managers have more information about a firm thanpotential debt or equity holders, make it difficult for potential creditorsand equity holders to evaluate the prospects of different firms. If creditorscannot distinguish between good quality and poor quality potential borrowersthen the market interest rate is likely to incorporate a premium – goodquality borrowers would be charged more than they would in a perfectly-informedmarket.[5]Similarly, new equity issues may trade at a discount to their value impliedby the underlying prospects of afirm.[6]The firm may also incur agency costs – costs borne by owners of the firmresulting from potential conflicts between managers, debtholders andequityholders.[7] For example,the nature of the debt contract may provide an incentive for managers, actingon behalf of equity holders, to pursue riskier investment projects than wouldbe pursued under a different financial structure. If the investment is successful,equity holders capture most of the excess gain; if the venture is unsuccessful,both equity holders and debt holders share the burden. Because debtholdersanticipate this type of behaviour, they price debt contracts accordingly.
The effect of these information problems is to boost the cost of external financerelative to internal finance. These cost differentials provide some insightinto how a given level of investment will be funded – cash flows willbe preferred to debt which, in turn, will be preferred to new equityissues.[8]Recent Australianevidence (Shuetrim, Lowe, and Morling (1993)) shows that the capitalstructure of the firm can be explained in part by these types of informationalproblems.
The theoretical extent of asymmetric information problems and agency costs can beshown to be a function of the structure of a firm's balance sheet. Accordingly,the structure of a firm's balance sheet will influence its investment decisionand shocks to the balance sheet will alter the evolution of investment overtime.[9]Firms can alter the cost of funding investment in a number of ways. Highercash flows directly reduce the cost of funds because firms will be less relianton more costly external funding. They also help reduce the costs of externalfunds by increasing the collateral backing of externalfinance.[10]Recent evidence from the United States suggests that firms often “reliquify”,that is, build up their stock of financial assets before undertaking largeinvestments (Whited (1991) and Eckstein and Sinai (1986)). They do this eitherbecause they have limited access to external finance or because it providesthem with collateral to obtain external funding at lower cost. Shifts in cashflows, financial assets and leverage may thus influence the dynamics of investment.Indeed, Bernanke and Gertler (1989) show that shocks to balance sheets canincrease the amplitude of the investment cycle in a simple neoclassical model.
Because the degree of asymmetric information and agency costs depends on firm characteristics,certain firms may be more sensitive to financial factors than others. For example,investors are likely to be less well-informed about smaller companies. Thismay hinder their ability to raise funds and boost the costs of external funding.Changes in cash flows may thus be a more important determinant of investmentfor smallercompanies.[11]Also, the investment of firms with higher leverage may be more sensitive tocash flows than that of firms with lower leverage. The increased debt servicingobligations resulting from higher leverage mean that the available cash flowsof higher-geared firms are smaller and thus they have less of a buffer againstdisturbances.
Consideration of these links between investment and the balance sheet position ofthe corporate sector enriches the theoretical representation of the way thatmonetary policy is transmitted. In simple models, monetary policy affects corporateinvestment directly by altering the rate at which the expected returns to investmentare discounted and indirectly through its effects on demand in the economygenerally. Adding financial factors into the analysis means that monetary policywill also affect investment through its effect on the financial position ofthe corporate sector. A tightening in policy, for example, will increase interestpayments and reduce cash flow. This will reduce the availability of relativelycheap internal funds and also increase the cost of external funds. Additionally,asset prices will fall, reducing the collateral that firms can provide to outsidefinanciers, raising the cost of external funding. Investment may be affectedthrough these channels in addition to the intertemporal substitution effectsof standard theory. Because the importance of these factors will vary acrossfirms depending on their size and financial structure, changes in monetarypolicy will be transmitted unevenly across the corporate sector.
Smaller firms have difficulty raising funds from capital markets for a variety ofreasons. For example, Woo and Lange (1992), note that “limited accessmay arise as a result of prohibitions or barriers to entry that specificallypreclude small firms from gaining funds, either through regulation or interms of the costs involved”.[3]
Oliner and Rudebusch (1989) find that transaction costs associated with externalfinance are significant in the US. For example, they report that transactioncosts account for up to one-quarter of the gross proceeds of small stockissues and one-seventh of the proceeds of small debt issues. Although currentcosts are likely to be considerably lower than this as a result of financialderegulation and innovation, these costs still remain important (see Allen(1991)).[4]
Akerlof (1970) and Stiglitz and Weiss (1981).[5]
Myers and Majluf (1984).[6]
There are many types of agency costs discussed in the literature (see Harris andRaviv (1991)).[7]
This financing hierarchy results because cash flows will be the cheapest source offunds, followed by debt and then by new equity. Debt will be cheaper thannew equity financing because the debt contract can be structured in sucha way as to minimise the consequences of the informational problems. Anumber of studies confirm the existence of financing hierarchies. ChaplinskyandNiehaus (1990) and Amihudet al. (1990), for example, find evidence that firms prefer internallysourced funds to external securities. Direct management surveys such asAllen (1991) and Pinegar and Wilbricht (1989) confirm these findings.[8]
Mills, Morling and Tease (1993) provide an analysis of the recent Australian experience.[9]
Bernanke and Gertler (1989) develop a model in which fluctuations in a firm'sbalance sheet change the agency costs of external funding and induce fluctuationsin investment. Agency costs are assumed to be positively related to collateralisablenet worth. This results in a cyclical relationship between balance sheetsand investment. During an upturn, for example, net worth increases, agencycosts are thus reduced and thus investment picks up. Similarly, shocksto net worth independent of the cycle will alter investment. See Lowe andRohling (1993) for Australian evidence.[10]
Gertler (1988) and Fazzari, Hubbard and Peterson (1988).[11]