Ebook Financing Shortfalls and the Value of Aggregate Liquidity
Liquid assets yield low returns, and tend to have higher prices and therefore lower expected returns when aggregate conditions are poor. We construct a model of the level and dynamics of the value of aggregate liquidity induced by firms’ financing shortfalls. Liquid assets are defined to be those which can be quickly reallocated at a low cost. In our model, liquidity and cash flows constitute internal funds available for investment in an economy where external funds are costly. The liquidity premium is then measured as the premium on funds which can be costlessly used for immediate investment. The use of liquidity to hedge investment opportunities can generate sizeable liquidity premia with the empirically observed cyclical properties. The model generates liquidity accumulation even if the expected return on liquid assets is much lower than that on real capital, and delivers a liquidity premium which is substantially countercyclical.
A lack of coincidence between sources and uses of funds is needed to generate corporate demand for liquid assets. The occurrence of such a lack of coincidence is not a foregone conclusion; in standard models the main use of funds (investment) is highly correlated with the main source of funds (cash flows). We construct a model where we break this strong correlation by introducing corporate finance frictions between the corporate and consumer sectors. Firms use funds to make discretionary and non-discretionary payments to investors, and to invest in productive investment opportunities. Likewise, firms have two sources of funds, internal funds, and new funds from consumers. Internal funds are the sum of current cash flow less committed payouts and the firm’s stock of liquid savings. Since raising external finance is costly, the value of liquidity is high when investment opportunities and available internal funds do not coincide. When such a mismatch occurs, the value of liquid assets inside the corporate sector where investment opportunities arise exceeds their intrinsic value. Liquid assets bear a convenience yield because they are fungible and this convenience yield varies with firms’ financing shortfalls. We study the resulting process for the value of liquid assets, and their expected return, and relate this value to measures of firms’ financing shortfalls.
Much of the existing theoretical literature studying the demand for liquid assets has focused on consumers/investors. However, it turns out to be difficult for models of consumers to generate a sizable liquidity premium. Consumers use buffer stocks of liquid assets to avoid selling illiquid assets, or costly borrowing. Moreover, in calibrated models consumers much prefer insuring negative shocks by saving out of them, and can effectively avoid selling even slightly illiquid assets. Firms may be a more important and quantitatively relevant source of liquidity demand for three reasons: First, firms are subject to larger shocks than consumers. Second, firms may try to limit internal funds because of agency problems, as in Jensen (1986) and Stulz (1990). Finally, and relatedly, firms rely more heavily on “external finance” than do consumers.
We build on models such as Holmström and Tirole (1996, 1998, 2001), which have developed important microeconomic foundations for corporate liquidity policies. Our model uses lessons from corporate finance to extend the canonical consumption based asset pricing model, or business cycle model to include a role for corporate liquidity demand.
To motivate our model, it is useful to consider the role of liquidity inside firms from the standpoint of standard asset pricing and macroeconomic models. Consider first the canonical consumption based asset pricing model with CRRA utility and a coefficient of relative risk aversion greater than one. In such models investment opportunities arrive as positive shocks to expected returns and are good news. When they arrive, investors are “richer” and can consume and/or invest more. With risk aversion greater than one, the income effect dominates the substitution effect and investors invest a lower fraction of income, and consume a higher fraction. Thus, investors actually want to consume relatively more instead of investing relatively more upon the arrival of investment opportunities, and there is never a mismatch between investment opportunities and available funds. Moreover, this model implies that investors should hedge states without investment opportunities. This runs counter to theories of hedging in corporate finance where firms instead desire to hedge states where investment opportunities arrive.
The difference between the two frameworks is that the corporate finance literature has focused more on the frictions between the agents who have funds, and the agents who have investment opportunities. Investment opportunities are no longer just “good news” if it is costly to take advantage of them. In other words, the fact that it may be costly to move funds to where investment opportunities reside alleviates the income effect of the arrival of such opportunities. This allows the substitution effect to dominate, leading consumers to invest a larger fraction of income rather than consuming it.
Next, consider the decentralized version of the standard real business cycle model. The decentralization works through complete markets, and hence there are no binding frictions between consumers and firms; Modigliani and Miller (1958) irrelevance holds in these economies. Thus, whether funds reside in the corporate or consumer sector is irrelevant. Moreover, due to consumption smoothing motives, consumers want to invest more in total (albeit a smaller fraction of income) when productivity is high. Since this is when current output is also high, the level of investment is high when funds for investment are plentiful. Thus, there is no role for corporate liquidity in the standard consumption based asset pricing model, or real business cycle model.
The basic intuition of the model works off of a tradeoff between two frictions which we introduce into the standard consumption based asset pricing model, or business cycle model. Frictions which make raising external finance costly lead firms to accumulate liquidity. We summarize such costs with the label “Myers” costs. However, firms with too much liquidity are subject to free cash flow agency problems, and this leads firms to pay funds out to consumers. We summarize these costs with the label “Jensen” costs. These two frictions de-link the value of liquid assets inside and outside the corporate sector, and distinguish the corporate from the consumer sector.
Our tradeoff between costly internal and external finance is very closely related to the tradeoff between under and over investment in Stulz (1990). His model illustrates how committed payouts can reduce the over investment problem, but lead to under investment. Riddick and Whited (2006) study the related question of the corporate propensity to save in the cross section. They study a tradeoff between costly external finance and tax costs of saving, and conduct a structural estimation of the parameters in the model using panel data. The tradeoff in their model is also closely related to ours, but the focus on comparative statics of corporate savings in the cross section over the variance of income shocks and costs of external finance is quite different. Both the theoretical and empirical results in their paper are complementary to ours.
The second modification we introduce is designed to de-link available funds and investment opportunities. We model investment opportunities as shocks to the price of new capital. If investment opportunities instead come from persistence in output productivity, then cash flows and investment opportunities are likely to coincide. Modeling investment opportunities in the form of lower new capital prices not only prevents the coincidence of available funds and investment opportunities, but also provides a counterbalance to the income effect discussed above by making the substitution effect towards higher investment stronger.
We analyze our model and study the level and variation in the value of liquidity as a function of the aggregate state. The model also provides implications for corporate payout policy, external finance activity, and corporate liquidity policy. We find that the value of aggregate liquidity is highest when investment opportunities arise and current productivity, and hence internal cash flow, or the “natural supply” of liquidity, is low. In these states, firms draw down their balances of liquid assets and raise external finance. Firms accumulate liquidity only when current output is high and there is no investment opportunity, and otherwise draw down their liquid funds.
Using the US Flow of Funds and Compustat data, we examine the time series properties of firms’ financing shortfalls. We construct time series of the shortfalls between aggregate internal funds and aggregate investment, and also compare our aggregate measure to a firm level measure that does not net out firm level shortfalls and surpluses. We compare the relationship between financing shortfalls and the value of aggregate liquidity in data generated from the model economy and US data. We expect the value of aggregate liquidity to be high when there is a shortfall in the aggregate corporate sector and find a positive relationship between our measure of shortfalls and the spread between commercial paper and treasury bills, a commonly used liquidity measure. Our measure of financing shortfalls also has predictive power for this spread. Section 2 describes our model and analytical and numerical results, section 3 describes our empirical findings, and section 4 concludes.
Download
PDF Ebook Financing Shortfalls and the Value of Aggregate Liquidity
Posted in :
